The extraction of high value chemicals
from heather (Calluna vulgaris) and
bracken (Pteridium aquilinum)
Jiewen Zhao
PhD
The University of York
Chemistry
April 2011
Abstract
The aim of the project is to extract and identify high value chemicals with potential
applications as pharmaceutical drugs or precursors from heather (Calluna vulgaris)
and bracken (Pteridium aquilinum). Supercritical carbon dioxide extracts of heather
were analysed for their triterpenoid content. Several triterpenoids were identified in
heather supercritical carbon dioxide extracts and many of them were already reported
to have potential valuable biological activities. Seasonal variation of total triterpenoid
content in heather was also evaluated. The result indicated that summer flowering
heather contains the highest concentration of total triterpenoid (15400 g/g dry plant)
and spring heather picked in March exhibited the second highest total triterpenoid
content (11200g/g dry plant). Optimal harvest and extraction time of heather is also
identified based on its triterpenoid seasonal variation trend.
In this study, supercritical carbon dioxide extraction of bracken was achieved for the
first time. Naturally occurring pterosin B and pterosin F were identified in bracken
supercritical CO2 extract and the seasonal variation in pterosin content and other
major compounds in bracken were also evaluated. Bracken crozier was found to
contain the highest content of pterosins, and the pterosin content reduced dramatically
when the frond grew towards its maturity. Geographical difference was also found to
cause significant variation in the level of pterosins. It was found that Welsh bracken
contains much less pterosin B and pterosin F compared to the Yorkshire sample. This
distinct difference may be due to different pH value of the soil in Kilburn (Yorkshire)
and Wales. Pterosins were recently proved to have strong anti-diabetic and
anti-obesity activities. Although pterosin B and F were not identified to have certain
anti-diabetic or anti-obesity activities in previous research, these pterosins, especially
pterosin F, have high possibility to be used as precursors to other more effective
pterosins because of its active chlorinated side chain.
List of Contents
List of tables .............................................................................................................. 9
List of illustrations ................................................................................................ 11
Acknowledgements ............................................................................................... 15
Declaration .............................................................................................................. 17
Heather and bracken as potential resources of high value chemicals ... 19
1.1
Project background......................................................................................... 20
1.2
Project aims .................................................................................................... 23
1.3
Green chemistry ............................................................................................. 24
1.4
Chemicals from heather and bracken ............................................................. 26
1.4.1
Natural chemicals from Calluna vulgaris ............................................... 26
1.4.1.1 Ursolic acid ...................................................................................... 27
1.4.1.2 Oleanolic acid................................................................................... 30
1.4.1.3 Lupeol .............................................................................................. 31
1.4.1.4 Phenolic compounds ........................................................................ 32
1.4.2
Natural chemicals from Pteridium aquilinum ......................................... 35
1.4.2.1 Ptaquiloside ...................................................................................... 35
1.4.2.2
1.5
Pterosins .......................................................................................... 37
Extraction and analysis methods .................................................................... 39
1.5.1
Supercritical carbon dioxide extraction .................................................. 39
1.5.2
Soxhlet extraction ................................................................................... 46
1.5.3
Supercritical fluid chromatography......................................................... 48
Supercritical carbon dioxide extractions of heather ................................. 53
2.1
Supercritical carbon dioxide extractions of heather ...................................... 54
2.2
General yield of supercritical carbon dioxide extractions ............................. 56
2.3
Column fractionation of heather supercritical CO2 and supercritical CO2 with
ethanol entrainer extracts and identification of novel triterpenoids .............. 58
2.3.1
Fractionation of heather supercritical CO2 extract .................................. 58
5
2.3.1.1 Molecular ion adjustment in GC-MS analysis .............................. 60
2.3.1.2 Identification of alkanes in heather supercritical CO2 extract
hexane fraction .............................................................................. 64
2.3.1.3 Quantification of triterpenoid compound ........................................ 68
2.3.1.4 Identification of novel triterpenoids in heather supercritical CO2
extract fractions ............................................................................. 68
2.3.1.5 Composition of triterpenoid compounds in heather supercritical
CO2 extract .................................................................................... 94
2.3.2
Fractionation of heather supercritical CO2 with ethanol entrainer extract
..................................................................................................................................... 95
2.3.2.1 Identification of alkanes in heather supercritical CO2 with entrainer
extract hexane fraction ......................................................................... 98
2.3.2.2 Identification
of
new
triterpenoid
compounds
in
heather
supercritical CO2 with ethanol entrainer extract fractions .......... 100
2.3.2.3 Composition of triterpenoid compounds in heather supercritical
CO2 with ethanol entrainer extract .............................................. 100
2.4
Conclusion and further remarks ................................................................... 102
Seasonal variation of total triterpenoids compounds in heather ........... 105
3.1
Seasonal variation of total triterpenoid compounds in heather.................... 106
3.2
Water content variation of heather............................................................... 106
3.3
Variation in total extract yield of Calluna vulgaris ..................................... 107
3.4
Seasonal variation of total triterpenoid content in heather .......................... 109
3.5
Comparison of total triterpenoid content extracted by supercritical carbon
dioxide and hexane ...................................................................................... 113
3.6
Optimal harvest and extraction time ............................................................ 115
3.7
Conclusion and further remarks ................................................................... 115
Extraction and quantification of heather total phenolic compounds .. 117
4.1
Extraction and quantification of heather phenolic compounds ................... 118
4.2
Seasonal variation of total phenolic compound in heather .......................... 119
4.3
Qualitative antimicrobial assay using heather phenolic extracts ................. 121
4.4
Quantitative antimicrobial test of heather flower phenolic extract.............. 126
4.5
Conclusion and further remarks ................................................................... 133
Liquid CO2 and supercritical CO2 extractions of bracken ...................... 137
5.1
Liquid CO2 and supercritical CO2 extractions of bracken ........................... 138
6
5.2
Liquid and supercritical carbon dioxide extraction of bracken crozier........ 139
5.2.1
Liquid CO2 extraction of bracken crozier ............................................. 139
5.2.1.1 Identification of pterosin B ............................................................ 140
5.2.1.2 Identification of pterosin F ............................................................. 144
5.2.1.3 Identification of other major compounds found in bracken crozier
liquid CO2 extract ........................................................................ 147
5.2.2
Supercritical carbon dioxide extractions of bracken crozier ................. 154
5.2.3
Supercritical carbon dioxide extraction of bracken crozier with ethanol
entrainer ............................................................................................... 160
5.3
Supercritical carbon dioxide extraction of bracken at different growth stages
................................................................................................................................. 162
5.3.1
Supercritical CO2 extractions of bracken samples at different growth
stages .................................................................................................... 162
5.3.2
Supercritical CO2 extractions of bracken with ethanol entrainer at
different growth stages.......................................................................... 165
5.4
Investigation of pterosin precursor in lady fern (Athyrium filix-femina) by
using supercritical fluid chromatography .................................................... 169
5.5
Conclusion and further remarks ................................................................... 175
Seasonal variation in ptaquiloside derivatives pterosins and other major
compounds.............................................................................................................. 179
6.1
Seasonal variation in ptaquiloside derivatives pterosins and other major
compounds ................................................................................................... 180
6.2
Water content variation of bracken .............................................................. 182
6.3
Variation in total extract yield of bracken over a life cycle ......................... 183
6.4
Seasonal variation of pterosins in bracken................................................... 185
6.5
Seasonal variation of -sitosterol................................................................. 189
6.6
Pterosins variation caused by geographical difference ................................ 192
6.7
Conclusion and further remarks ................................................................... 197
Materials, methods and experimental procedures ...................................... 199
7.1
Materials and reagents ................................................................................. 200
7.2
Harvesting and milling process of heather and bracken .............................. 200
7.2.1
Harvesting and preparation of heather material .................................... 200
7.2.2
Harvesting and preparation of bracken ................................................. 202
7.2.3
Water content of heather ....................................................................... 202
7.3
Extraction and fractionation of heather ........................................................ 203
7
7.3.1
Soxhlet extractions of heather............................................................... 203
7.3.2
Supercritical carbon dioxide extraction of heather ............................... 203
7.3.3
Fractionation of crude heather extract using column chromatography 204
7.3.4
Thin layer chromatography (TLC) of crude heather extract and fractions
obtained after column chromatography ................................................ 204
7.3.5
Anti-microbial assays using heather extracts ....................................... 205
7.3.5.1 Sample preparation for paper disc antimicrobial test (qualitative test)
................................................................................................................................... 205
7.3.5.2 Preparation for multi-well plate antimicrobial test (quantitative test)
................................................................................................................................... 205
7.3.6 UV analysis of heather total phenolic content ........................................ 205
7.3.7 UV analysis of heather total triterpenoid content ................................... 206
7.3.8 High temperature-Gas chromatography mass spectrometry (HT-GCMS)
of heather extracts ................................................................................... 207
7.4
Extraction and fractionation of bracken....................................................... 207
7.4.1 Supercritical carbon dioxide extraction of bracken fern......................... 207
7.4.2 High temperature-Gas chromatography mass spectrometry (HT-GCMS)
method for bracken analysis ................................................................... 208
7.4.3 Methanol extraction of bracken for supercritical fluid chromatography 209
7.4.4 Supercritical Fluid Chromatography of bracken extracts ....................... 209
Conclusion and further work ........................................................................... 211
8.1
Concluding remarks ..................................................................................... 212
8.1.1 General background ................................................................................ 212
8.1.2 Conclusions from heather study ............................................................. 213
8.1.3 Conclusions from bracken study ............................................................ 215
8.2
Future work and publications ...................................................................... 219
List of references ................................................................................................. 221
8
List of tables
1.1
The 12 fundamental principle of Green Chemistry ...................................... 25
1.2
Representative example of distribution of ursolic acid in folk medicines and
its pharmacological activities ......................................................................... 29
1.3
Seasonal variation in phenolic compounds of Calluna vulgaris shoots and
roots................................................................................................................ 32
1.4
Comparison of physical and transport properties of gases, liquids, and SCF ....
................................................................................................................................... 40
1.5
Co-solvents used to modify carbon dioxide for SFE ..................................... 43
1.6
Selected solubility of organic compounds in liquid CO2 ............................... 45
1.7
The advantages and disadvantages of conventional Soxhlet extraction ........ 48
2.1
Abundances (in g/g dry plant) of triterpenoids in leaves, stems and roots of
Calluna vulgaris and Erica tetralix ............................................................... 55
2.2
Triterpenoid compounds found in heather supercritical CO2 extract fractions
........................................................................................................................ 69
2.3
Triterpenoid compounds found in heather supercritical CO2 with ethanol
entrainer extract fractions ............................................................................ 100
3.1
Calculated total triterpenoid content in each heather supercritical extract ... 111
4.1
Extract yield and total phenolic content in each heather sample .................. 120
4.2
Antibacterial activity of heather phenolic extracts against common
Gram-positive and Gram-negative bacteria ................................................. 124
4.3
The arrangement of the plate ........................................................................ 130
4.4
Turbidity readings at 485nm of each well after incubation .......................... 131
4.5
Average turbidity readings of each column and calculated inhibition effect of
phenolic extracts .......................................................................................... 132
5.1
Kovats index calculated for the two peaks and comparison with published
data ............................................................................................................... 154
5.2
Calculated KI value of alcohol peak series and corresponding standard KI
value ............................................................................................................. 159
5.3
The extract yields of supercritical carbon dioxide extractions ..................... 163
5.4
The extract yields of supercritical carbon dioxide extractions with ethanol
entrainer ....................................................................................................... 166
6.1
Extract yield of supercritical carbon dioxide extractions ............................. 183
9
10
List of illustrations
1.1
Calluna vulgaris ............................................................................................. 22
1.2
Pteridium aquilinum....................................................................................... 22
1.3
Chemical structure of ursolic acid ................................................................. 27
1.4
Chemical structure of oleanolic acid .............................................................. 30
1.5
Chemical structure of lupeol .......................................................................... 31
1.6
Chemical structure of chlorogenic acid.......................................................... 33
1.7
Chemical structure of quercetin 3-O-glucoside ............................................. 33
1.8
Chemical structure of quercetin 3-O-galactoside ........................................... 34
1.9
Chemical structure of quercetin 3-O-arabinoside .......................................... 34
1.10
Chemical structure of (+)-catechin................................................................. 34
1.11
Chemical structure of ptaquiloside ................................................................ 35
1.12
Ptaquiloside reactions..................................................................................... 36
1.13
Chemical structures of some bracken pterosins ............................................. 37
1.14
Phase diagram for supercritical CO2 .............................................................. 40
1.15
Variation of density with pressure for carbon dioxide ................................... 41
1.16
A detailed diagram of supercritical CO2 extraction apparatus ....................... 42
1.17
A schematic diagram of soxhlet extraction device ........................................ 47
1.18
Semi-prep scale supercritical fluid chromatography in Green Chemistry
group, University of York .............................................................................. 49
1.19
Polarity comparison between modified CO2 with common LC solvents ...... 50
2.1
Numbering of carbon skeleton of triterpenoids .............................................. 54
2.2
Extraction yield from dried heather using supercritical CO2 and supercritical
CO2 with ethanol entrainer............................................................................. 57
2.3
Extraction yield from fresh heather using supercritical CO2 and supercritical
CO2 with ethanol entrainer............................................................................. 57
2.4
TLC plate with separated spots revealed from the column of supercritical CO2
extract ............................................................................................................. 59
2.5
Mass spectrum of squalene obtained from Perkin Elmer Clarus 560S mass
spectrometer compared with NIST library standard ...................................... 61
2.6
Mass spectrum of -amyrin obtained from Perkin Elmer Clarus 560S mass
spectrometer compared with NIST library .................................................... 62
2.7
Mass spectrum of stigmasterol obtained from Perkin Elmer Clarus 560S mass
spectrometer compared with NIST library standard ...................................... 63
2.8
Gas chromatogram of fraction 1 compared with C12-50 hydrocarbon
11
standards ........................................................................................................ 65
2.9
Mass spectrum of peak at 28.94min and standard spectrum of squalene
obtained from NIST library ........................................................................... 67
2.10
The standard curve of oleanolic acid for heather triterpenoid compound
quantification ................................................................................................. 68
2.11
Gas chromatogram of fraction 2 and fraction 3 with triterpenoid group
highlighted ..................................................................................................... 70
2.12 Gas chromatogram of fractions 4, 5, and 6 with triterpenoid group highlighted
....................................................................................................................... 71
2.13 Expanded gas chromatogram of fraction 2 with each identified triterpenoid
compound highlighted ................................................................................... 72
2.14 Chemical structure of taraxerone ................................................................... 73
2.15 Mass spectrum of the peak at 32.96min in fraction 2 compared with standard
taraxerone spectrum ....................................................................................... 75
2.16 Chemical structure of taraxerol ...................................................................... 76
2.17 Mass spectrum of peak at 32.93min in fraction 3 compared with standard
taraxerol spectrum ......................................................................................... 79
2.18 The mass spectrum of the peak at 34.63min in fraction 2 compared with
standard spectrum of friedelin ....................................................................... 83
2.19 Chemical structure of 13,27-cycloursan-3-one .............................................. 84
2.20 The mass spectrum of the peak at 33.28min in fraction 2 compared with
standard spectrum of 13,27-cycloursan-3-one............................................... 85
2.21 The mass spectrum of the peak at 33.60 min in fraction 3 compared with
standard spectrum of 9-methyl-19-Norlanosta-5,24-dien-3-ol in previous
research .......................................................................................................... 88
2.22 The tandard mass spectrum of 9-methyl-19-Norlanosta-5,24-dien-3-ol in Nist
library............................................................................................................. 89
2.23 The mass spectrum of the peak at 33.52 min in fraction 3 compared with
standard spectrum of -amyrin...................................................................... 92
2.24 The mass spectrum of the peak at 33.17 min in fraction 3 compared with
standard spectrum of -amyrin ...................................................................... 93
2.25 Triterpenoid distribution of heather supercritical CO2 extract ....................... 95
2.26 TLC plate (1) with spots revealed for supercritical CO2 with entrainer extract
....................................................................................................................... 96
2.27 TLC plate (2) with spots revealed for supercritical CO2 with entrainer extract
....................................................................................................................... 97
12
2.28
Gas chromatogram of fraction 1 of heather supercritical CO2 with ethanol
entrainer extract.............................................................................................. 99
2.29
Triterpenoid distribution of heather supercritical CO2 with ethanol entrainer
extract ........................................................................................................... 101
3.1
Seasonal variation in water content of Calluna vulgaris .............................. 107
3.2
Comparison of hexane extractions and supercritical extractions in dry plant
...................................................................................................................... 108
3.3
Comparison of hexane extractions and supercritical extractions in fresh plant
...................................................................................................................... 109
3.4
The standard curve of oleanolic acid for heather total triterpenoid
quantification ............................................................................................... 110
3.5
Total triterpenoids in supercritical CO2 and supercritical CO2 with ethanol
entrainer extracts (dry plant) ...................................................................... 111
3.6
Total triterpenoids in supercritical CO2 and supercritical CO2 with ethanol
entrainer extracts (fresh plant) ................................................................... 112
3.7
Total triterpenoids in supercritical CO2 and heather hexane extracts (dry plant)
...................................................................................................................... 113
3.8
Total triterpenoids in supercritical CO2 and heather hexane extracts (fresh
plant) .......................................................................................................... 114
4.1
The standard curve of gallic acid for heather total phenolics quantification 119
4.2
Seasonal variation of total phenolic content in heather ................................ 120
4.3
Effect of 10/07 flower, 10/07 whole plant, ethanol control and water control
on S. aureus .................................................................................................. 122
4.4
Effect of 1/08, 3/08, 8/08, ethanol control and water control on S. aureus .. 122
4.5
Effect of 1/08, 8/08, 10/07whole plant, ethanol control and water control on E.
coli................................................................................................................ 123
4.6
Effect of 10/07 flower, 3/08, ethanol control and water control on E. coli .. 123
4.7
Seasonal variation of antimicrobial activity of heather phenolic extracts .... 125
4.8
Round bottomed 96 multi-well culture plate ................................................ 127
4.9
The bacteria colonies of pH 4-pH 8 heather phenolic extracts ..................... 128
4.10
The bacteria colonies of pH 4-pH 8 water control ....................................... 128
5.1
Gas chromatogram of bracken crozier liquid CO2 extract............................ 141
5.2
Mass spectrum of pterosin B ........................................................................ 142
5.3
GC-MS data of pterosin B published by Bonadies ....................................... 142
5.4
Mass spectrum of pterosin F ......................................................................... 145
5.5
Mass spectra of pterosin F published by Bonadies ....................................... 145
5.6
Mass spectra of -sitosterol and the peak with retention time at 33.42min . 150
13
5.7
Chemical structures of stigmasterol and -sitosterol ................................... 151
5.8
The standard curve of stigmasterol for -sitosterol quantification .............. 152
5.9
Mass spectra of the two peaks in liquid CO2 extract and 1-triacontanol
standard ........................................................................................................ 153
5.10 Gas chromatogram of bracken crozier supercritical CO2 extract ................. 156
5.11 Mass spectra of long chain alcohol peaks in supercritical CO2 extract and
1-triacontanol standard ................................................................................ 158
5.12 Gas chromatogram of bracken crozier supercritical CO2 with ethanol extract
..................................................................................................................... 161
5.13 Gas chromatogram of bracken supercritical CO2 extracts ........................... 164
5.14 Gas chromatogram of bracken supercritical CO2 with EtOH extracts ......... 168
5.15 Gas chromatogram of lady fern ethanol extract ........................................... 171
5.16 Supercritical chromatogram of bracken fern and lady fern ethanol extracts 172
5.17 Gas chromatogram and mass spectra of the fraction and gas chromatogram of
C12-60 alkane standard ............................................................................... 174
6.1
The crozier step in bracken vegetation process ............................................ 180
6.2
2nd step in bracken’s life cycle-with first pair of leaves ............................... 181
6.3
3rd step in bracken’s life cycle-mature plant................................................. 181
6.4
4th step in bracken’s life cycle-dying ............................................................ 181
6.5
Seasonal variation in water content of bracken fern .................................... 182
6.6
Seasonal variation in supercritical extraction yield and total yield of dry plant
..................................................................................................................... 184
6.7
Seasonal variation in supercritical extract yield and total yield of fresh
bracken ......................................................................................................... 185
6.8
Original GC traces of seasonal bracken supercritical CO2 extract ............... 186
6.9
GC traces of seasonal bracken supercritical CO2 extract with octadecane
standard ........................................................................................................ 186
6.10 The standard curve of stigmasterol for -sitosterol quantification .............. 190
6.11 Seasonal variation of -sitosterol in bracken dry plant ................................ 191
6.12 Seasonal variation of -sitosterol in bracken fresh plant ............................. 192
6.13 Comparison of Kilburn and Wales bracken supercritical CO2 extract ......... 194
6.14 Map of soil pH in the UK ............................................................................. 195
7.1
Milled whole heather .................................................................................... 201
7.2
Flowering heather ......................................................................................... 201
7.3
Separated heather flowers ............................................................................ 201
14
Acknowledgments
Firstly, I would like to make a special thank you to Professor Ray Marriott for all his
generous help and guidance. I am very grateful that I could be studying under his
supervision for two years, and I really appreciate all the encouragement and
inspiration he’s given to me. I also would like to thank Professor James Clark for his
supervision and giving me the opportunity to do this PhD.
I would like to thank Mr. Wild and the department for offering me Wild Fund
scholarship. I also would like to thank Karl Heaton for helping me doing mass
spectrum analysis and Adrian Harrison from biology department for providing
bacteria strains and helping me doing anti-microbial tests.
Huge thanks to Emily, Brigid, Nontipa, Alice, Peter, Gareth and Andy for all your
help in the lab and in the office. Thanks to Ray, James and Andy for proof reading my
thesis. Thanks to Jenny Wall for taking me to the moors for heather harvest in my first
year of study. A big thank you also needs to be given to all members of the group for
your help with work.
Finally, I would like to thank my parents and Mr. Guo for their unconditional support
and consistent understanding.
15
16
Declaration
Some of the results presented in this thesis were obtained by, or in collaboration with
other workers, who are fully acknowledged in the text. All the other results are the
work of the author.
April 2011
17
18
Chapter 1
Introduction
Heather and bracken as potential
resource of high value chemicals
Chapter 1
19
Chapter 1
Introduction
1.1 Project background
It is acknowledged that moorland habitats dominated by heather (Calluna vulgaris)
and related undershrubs are of high conservation value within the European Union
and especially the United Kingdom.1 British moorland represents 20% of the
European resource of this high conservation value community.
British upland moorlands are usually designated as National Parks for the sake of
protecting vegetation and fauna. The North York Moors National Park contains the
largest patch of continuous heather moorland (50,100 ha) in England, representing
over 10% of the country’s resource.1 Within the National Park 88% (44,088 ha) of
moorland habitat is designated as a Site of Special Scientific Interest (SSSI), Special
Protection Area (SPA) and Special Area for Conservation (SAC). Dry heath
vegetation covers 26,500 ha and forms the main land cover on the western, southern
and central moors. Heather or ling is the dominant species over this entire area with
patches of bell heather. Heather is also abundant on the wet heath (8,700 ha), which
occurs on moister soils. The most noticeable value of heather is as bird habitats for
breeding and feeding. To suit the bird’s different phases of growth, moorland needs to
be cut to various heights. Burning, alongside grazing, is still the predominant method
of managing moorland vegetation.2 If more value can be added to this undeveloped
asset, then the farmers in upland areas or the national parks authority could harvest it
rather than burn it. The extraction of chemicals from such plants by environmental
benign techniques would hopefully give a new life to this generally considered waste
resource.
As well as heather, bracken fern (Pteridium aquilinum) is a widely distributing
species which dominates the vegetation of large areas in many UK upland regions.
The amount of land dominated by bracken range from 2880 km2, which is
20
Chapter 1
Introduction
approximately 1.3% of the land area of Great Britain, to 6361 km2, 2.8% of the land
area.3 The total area of bracken is estimated at 12,300 ha (24.6%) in North York
Moors National Park. Bracken is a major weed problem in Britain, particularly in the
uplands of the west and north. The major ecological effect of an expansion of bracken
is the loss of large areas of semi-natural vegetation. The loss of heather moorland
through bracken invasion is viewed as disastrous especially for the upland fauna
which is already affected by fragmentation of moorland habitat. Moreover, the
continuing spread of bracken can be seen as a threat to farming in the uplands as well.
The increasing area under bracken could lead to overgrazing elsewhere, which could
help the spread of bracken and also accelerate moorland degradation to grass.
Bracken fern is a perennial fern, it’s also one of the higher plants known to cause
cancer naturally in animals.4 In the past it was an important rural resource and was
used for animal bedding, thatch, potash, soap and compost. Since the rural and
industrial applications declined, it no longer regarded as an asset.5 Meanwhile, a
number of well-recognized toxic effects on livestock have been known since the end
of the 19th century. In cattle, it can cause a syndrome named cattle bracken poisoning.
It can induce urinary bladder cancer, and can also initiate thiamine deficiency in pigs
and horses, haemorrhagic syndrome in cattle, bright blindness in sheep and upper
gastrointestinal tracts tumour in all ruminants.4
Currently the use of the herbicide Asulam (methyl N-[(4-aminophenyl)sulfonyl]
carbamate) is most widely used for bracken control.3 It has a low toxicity for animals
and affects few other plants; however it still has some disadvantages such as high cost
and spray drift which can affect populations of other ferns. Bracken can also be
controlled by mechanical means such as cutting or rolling. This method can be
well-targeted at bracken fronds to prevent spread, however it requires to be carried out
regularly, e.g. twice a year for 12 years, to gives good control but not elimination.3 If
21
Chapter 1
Introduction
any valuable compounds can be extracted from bracken with a potential to be applied
as pharmaceutical drugs or precursors, more value would be added to this plant which
is currently viewed as a problem.
The major toxin and carcinogen has been shown to be ptaquiloside, a
nor-sesquiterpene glycoside which can be present in bracken at high concentration.6
Its hydrolysate pterosins are a large group of biologically active sesquiterpenes.
Although none of the pterosins were found to be carcinogenic, some of them were
reported to be cytotoxic to HeLa cells.7 Other properties such as antimicrobial8 and
smooth muscle relaxant activity9 of some pterosins were also identified in the last
decade. Because of these properties and the link between pterosins and bracken toxin,
synthetic methods of different pterosins were widely researched. Since pterosins can
be easily converted from the unstable precursor ptaquiloside by mild acid or base
treatment, from a green chemistry perspective it is better to develop a conversion
route from ptaquiloside to certain pterosins rather than complex synthesis using large
quantities of chemicals and catalysts. The high content of ptaquiloside may afford a
large scale of preparation of the pterosins without producing much chemical waste.
Figure 1.1 Calluna vulgaris
Figure 1.2 Pteridium aquilinum
22
Chapter 1
Introduction
1.2 Project Aims
The core aim and focus of this project was to extract and characterise natural
chemicals with high value from heather (Calluna vulgaris) and bracken (Pteridium
aquilinum) using green chemical technologies that can be used directly as bioactive
molecules or as precursors for new pharmaceuticals. These technologies used include
supercritical fluid extraction in combination with other benign solvents. The extracts
were analysed using a number of analytical techniques such as GC, GC-MS, HPLC,
LC-MS and Supercritical Fluid Chromatography.
Summary of aims:
 Extraction of high value chemicals with biological activities such as triterpenoids
from heather using supercritical carbon dioxide extraction and conventional green
solvent extraction.
 Extraction of pterosins with potential pharmaceutical applications from bracken
and analysis of conversion routes to specific molecules.
 Investigate environmental benign co-solvents to enhance the selectivity of CO2 to
polar components.
 Isolate, purify and identify triterpenoids and other high value natural chemicals
from crude extracts, and also identify applications for these molecules.
 Examine the seasonal variation in heather triterpenoids content and bracken
pterosins content.
 Improve conventional methods of analysis by using a greener solvent.
23
Chapter 1
Introduction
1.3 Green chemistry
Green Chemistry is the design, development, and implementation of chemical
products and processes to decrease or eliminate the use and production of substances
harmful to human health and the environment. Green Chemistry is a basic
fundamental way to implement sustainable development and the application of the
Twelve Principles of Green Chemistry has proved that it is possible to arrive at a
compromise between society and economy, resources and environment, equity and
efficiency, mankind and nature by designing in sustainability at the molecular level.10
Table 1.1 shows the 12 Principles of Green Chemistry. Application of the 12
fundamental principles of Green Chemistry can achieve higher efficiency and less
environmental pressure during chemical synthesis.
24
Chapter 1
Introduction
Table 1.1： The 12 fundamental principle of Green Chemistry11
12 Fundamental principles
1
2
3
4
5
6
7
8
9
10
11
12
It is better to prevent waste than to treat or clean up waste after it is
formed.
Synthetic methods should be designed to maximize the incorporation of all
materials used in the process into the final product.
Wherever practicable, synthetic methodologies should be designed to use
and generate substances that possess little or no toxicity to human health
and the environment.
Chemical products should be designed to preserve efficacy of function
while reducing toxicity.
The use of auxiliary substances should be made unnecessary wherever
possible and, innocuous when used.
Energy requirements should be recognized for their environmental and
economic impacts and should be minimized. Synthetic methods should be
conducted at ambient temperature and pressure.
A raw material of feedstock should be renewable rather than depleting
wherever technically and economically practicable.
Unnecessary derivatisation should be avoided whenever possible.
Catalytic reagents are superior to stoichiometric reagents.
Chemical products should be designed so that at the end of their function
they do not persist in the environment and break down into innocuous
degradation products.
Analytical methodologies need to be further developed to allow for
real-time, in-process monitoring and control prior to the formation of
hazardous substances.
Substances and the form of a substance used in a chemical process should
be chosen so as to minimize the potential for chemical accidents.
According to the 7th principle, the use of renewable raw materials is of great
importance to green chemistry practice and there is growing urgency to develop
bio-based products produced from renewable resource. Bio-derived material is made
from renewable agricultural and forestry feed stocks including wood, wood waste and
residues, grasses, crop and crop by-product. Using bio-derived material as a
renewable resource is not only a solution to growing environmental threat but also a
solution to the uncertainty of petroleum supply.12
25
Chapter 1
Introduction
The extraction of natural wax and lipid from plants normally requires large amount of
solvent,13 but with the application of supercritical carbon dioxide extraction, the
solvent usage and extraction time can be dramatically reduced. In addition,
supercritical CO2 extraction also shows preferable selectivity in some cases.2
Moreover, the recent development of analytical techniques such as supercritical fluid
chromatography has brought quicker analysis and better selectivity into practice. In a
word, for a sustainable future of chemistry, all the processes must reach equilibrium
for the highest efficiency, safety and economy.
1.4 Chemicals from heather and bracken
Heather and bracken are excellent feedstocks for natural extracts with a wide range of
biological activities. The existence of several groups of natural chemicals with
anticancer, antioxidant and antimicrobial activities has been identified in heather.
Bracken contains a major toxin, ptaquiloside, which is well known for its
carcinogenicity. However antitumor activity and cytotoxicity to some cancer cells has
also been reported.
1.4.1 Natural chemicals from Calluna Vulgaris
The confirmation of the biological activities of triterpenoids has gained more and
more attention. Triterpenoids are ubiquitous throughout the plant kingdom in the form
of free acids or aglycones of saponins and exhibit various biological activities.14
Already, more than 80 triterpenoids have been identified from plants, and the number
of papers describing their bioactive effects has increased sharply during the last
decade.15 Specifically some of them have already been used as anticancer and
anti-inflammatory agents in Asian countries.16
26
Chapter 1
Introduction
At the same time, heather is also rich in phenolic acids, which are well known for
their antioxidant ability by donating hydrogen or electrons. Meanwhile, their stable
radical intermediates can prevent the oxidation of many food ingredients, especially
fatty acids and oils.17 Lambropoulos and co-workers18 proved that red wine phenolic
extracts at 100mg/L inhibited the oxidation of corn oil stripped of tocopherols to a
greater extent than butylated hydroxyanisole at 200 mg/L. Previous research has also
indicated that phenolic compounds have antimutagenic, anticarcinogenic, and
antiglycemic beneficial properties. To utilize these properties, phenolic compounds
can be added to health promoting food to prevent potential chronic diseases.19
1.4.1.1 Ursolic acid
Ursolic acid (3-hydroxy-urs-12-en-28-oic acid) is a natural pentacyclic triterpenoid
carboxylic acid and is the major component of some traditional medicine herbs. It is
well known for its biological effects, for instance antioxidation, anti-inflammation,
and anticancer activities. In recent years it has attracted considerable attention because
of its pharmacological effects combined with a relatively low toxicity.16
COOH
HO
Figure1.3: Chemical structure of ursolic acid
Ursolic acid exists naturally in many plants species including fruits and herbs (Table
1.2). Conventional applications of plants containing ursolic acid in folk medicine are
27
Chapter 1
Introduction
plentiful. For instance, the leaf and shoot tips of the Labrador tea plant (Ledum
groenlandicum Retzius), an Ericaceae species widely spread throughout North
America, are used in traditional American medicine to fight against some
inflammatory diseases such as asthma, rheumatism, and also diseases of the liver and
kidney. Lately, several instances of medicative efficacy of plants containing ursolic
acid have been shown and this may contribute to the further use of plants that contain
this chemical in both folk medicine and clinical research.15
More and more reports indicate that some natural plant compounds have conspicuous
inflammatory and cancer preventive effects.20,21 Ursolic acid is regarded as one such
compound since it possesses many biological activities, such as anti-oxidation,
anti-inflammatory, anticancer, and hepato-protection, as well as the ability to induce
apoptosis. In the early 1990s, the antioxidative activities of ursolic acid against free
radical-induced damage were studied.22, 23 In these experiments, ursolic acid showed
remarkable anti-oxidative activities by scavenging free radicals to four standard
chemical (ascorbic acid, carbon tetrachloride, ADP/iron, and adriamycin) induced
lipid peroxidation in isolated rat liver and heart microsome in vitro.
28
Chapter 1
Introduction
Table 1.2: Representative example of distribution of ursolic acid in folk medicines
and its pharmacological activities15
Plant name
Botanical name
Biological activity
Apple
Malus pumila
Anti-proliferation
Anti-cancer
Basil
Ocimum basilicum
Antiviral
Blueberry
Vaccinium spp.
Anti-cancer
Cranberry
Vaccinium macrocarpon
Anti-cancer
Ground ivy
Glechoma hederacea L.
Anti-cancer
Guava
Psidium guajava
Unknown
Heather flower
Calluna vulgaris
Anti-inflammatory
Japanese cherry
Prunus serrulata var. spontanea
Unknown
Labrador tea
Ledum groenlandicum Retzius
Antioxidant
Anti-inflammatory
Anti-cancer
Loquat
Eribotrya japonica Lindl.
Anti-mutagenic
Olive
Olea europaea
Antioxidant
Anti-atherosclerotic
Anti-hypertensive
Oregano
Origanum vulgare
Anti-leukemic
Antioxidant
Persimmon
Diospyros leucomelas
Antioxidant
Anti-inflammatory
Plantain
Plantago major L.
Antioxidant
Anti-inflammatory
Rosemary
Rosmarinus officinalis L.
Anti-inflammatory
Anti-cancer
Sage
Salvia officinalis L.
Anti-inflammatory
Thyme
Thymus
Anti-inflammatory
29
Chapter 1
Introduction
1.4.1.2 Oleanolic acid
Oleanolic acid (3b-hydroxyolean-12-en-28-oic acid) exists in ginseng root at a high
concentration. Along with ursolic acid, oleanolic acid has been found in more than
120 plants in the form of free acid or aglycones of the saponin group of
triterpenoids.20 Plants that contain oleanolic acid have been used as a drug for
longevity in East Asian countries since ancient times.15
COOH
HO
Figure1.4: Chemical structure of oleanolic acid
Oleanolic acid has been used for treatment of non-lymphatic leukemia24 and has been
shown to have the same biological activities as ursolic acid, including antioxidative,
22,23
anti-inflammation,25-27 anticancer,28-30 and anti-hepatotoxic31-33 activities. Among
these biological effects, the protection it provides to the liver against acute chemical
induced liver damage and chronic liver fibrosis and cirrhosis is the most widely
known. Oleanolic acid has been used individually or combined with other
hepato-protective components as an oral drug.14 The effective function of this
triterpenoid for combating liver diseases could be due to its anti-oxidant and
anti-inflammatory effects, and their actions on drug-metabolising enzymes. Oleanolic
acid is an efficient revulsant of metallothionein, a small cysteine-rich protein acting
like glutathione in the body’s defence against toxin invasion.14
30
Chapter 1
Introduction
As well as ursolic acid, oleanolic acid has various anticarcinogenic effects including
inhibiting tumour genesis and development, inducing cancer cell differentiation and
apoptosis.14 Oleanolic acid and its derivatives are also effective in inhibiting
angiogenesis, tumour cells infiltration and metastasis, and more and more are used as
a new class of chemotherapeutics.34 The mechanisms of the anticancer effects by
oleanolic acid need further investigation.
1.4.1.3 Lupeol
Lupeol (lup-20(29)-en-3b-ol), is a naturally occurring triterpene found in a range of
edible fruits such as olive, mango, strawberry, red grape, fig, and some medicinal
herbs, and is also widely used as a folk medicine in the Far East, North America,
Latin America, and the Caribbean islands.35-37 Lupeol and some analogues have many
proven biological activities such as antioxidative, anti-inflammatory and anticancer.38
Since they are widely distributed in abundant plants, lupeol and its analogues are
easier to obtain than other pharmaceuticals that are being used at the moment,
therefore make them a good potential herbal medicine to prevent cancer, coronary and
hepatic diseases. Moreover, it’s also revealed that lupeol showed low cytotoxicity on
healthy cells and can act as synergist in combined therapies, which makes it worth
investigating for single use or as an auxiliary drug to complement anti-inflammatory,
antineoplatic and anti-hypertensive therapies that are already used.38
Figure 1.5: Chemical structure of lupeol
31
Chapter 1
Introduction
1.4.1.4 Phenolic compounds
Phenolics are one of the largest and most widespread groups of plant secondary
compounds with multiple bioactivities such as antioxidation, antibiotic, anti-ulcer, and
anti-inflammatory. Polyphenols also have many important industrial applications such
as in the production of paints, paper and cosmetics, as tanning agents and in the food
industry as additives. Previous researchers indicated that Calluna vulgaris tissues
have a high total phenolic content, moreover, quantitative and qualitative
determination of phenolic composition and seasonal variation of phenolic content are
well understood.39
Table 1.3: Seasonal variation in phenolic compounds of Calluna vulgaris shoots and roots39
Jan
Feb Mar Apr
Ma
Jun
Jul
y
Au
Sep Oct
g
No
De
v
c
Caffeic acid
-
1
2
2
1
-
-
-
-
-
-
-
Orcinol
-
-
2
1
-
-
-
-
-
-
-
-
Orcinol--D-gluco
-
--
-
-
2
2
3
1
2
2
3
2
Arbutin
1
-
-
-
1
-
1
-
-
1
1
2
Chlorogenic acid
2
3
1
2
4
3
3
3
3
3
2
4
Isochlorogenic
-
-
1
2
-
-
-
-
-
-
-
-
Callunin
-
-
-
2
4
3
-
4
2
2
-
2
Kaempherol
-
-
1
1
1
1
1
-
1
-
1
1
Quercetin
1
1
1
1
2
3
3
1
4
1
4
2
Kaemphherol
-
-
-
1
1
1
1
1
1
1
1
-
-
-
-
1
1
1
1
1
1
1
1
-
side
acid
3-O-glucoside
Kaemphherol-3-Ogalacoside
32
Chapter 1
Introduction
Kaemphherol
-
-
-
1
1
1
1
1
-
-
1
-
4
1
1
2
3
4
3
4
4
3
3
3
4
1
1
2
3
4
3
4
4
3
3
3
1
1
1
1
2
2
1
1
2
2
2
1
(+)-Catechin
1
1
-
1
2
3
3
3
2
3
4
3
(-)-Epicatechin
1
1
-
1
1
1
1
1
1
1
2
1
Procyanidin D1
1
1
2
2
2
3
2
-
1
3
2
2
Procyanidin B1
-
1
1
1
1
1
1
-
-
1
1
1
Procyanidin B2
-
-
-
-
1
-
1
1
1
1
1
1
Procyanidin B3
-
-
-
1
1
1
1
-
1
-
-
-
ProcyanidinB4
-
-
-
-
-
1
1
-
-
-
-
-
Procyanidin B5
-
-
-
-
-
1
1
-
1
-
-
-
Procyanidin C1
-
1
1
-
1
1
1
1
1
1
1
1
Total phenolic
16.0
15.6
15.3
16.2
23.4
21.7
26.5
19.4
28.4
23.0
21.9
19.8
3-O-arabinoside
Quercetin
3-O-glucoside
Quercetin
3-O-galacoside
Quercetin
3-O-arabinoside
content in shoots
(%dry wt)
Figure 1.6: Chemical structure
of chlorogenic acid
Figure 1.7: Chemical structure of
quercetin 3-O-glucoside
33
Chapter 1
Introduction
Figure 1.8: Chemical structure
of quercetin 3-O-galactoside
Figure 1.9: Chemical structure
of quercetin 3-O-arabinoside
Figure 1.10: Chemical
structure of (+)-catechin
It is well known that phenolic compounds are powerful antioxidants, it has attracted
increasing interest in the food industry because they can slow oxidative degradation of
lipids and thereby improve the quality and nutritional value of food.40 In addition, all
aerobic life suffers oxidative pressure from reactive oxygen species (ROS), excess
ROS can lead to a series of pathological changes through lipid peroxidation and
protein and nucleic acid damage, and these changes may eventually induce chronic
diseases including cancer, cardiovascular disease, aging and neurodegenerative
disorders.41 Thus dietary phenolic antioxidants have been paid increasingly attention
as preventive ingredient.42
34
Chapter 1
Introduction
As well as antioxidation, the antimicrobial activity of phenolic compounds is also
significant. Corrales and co-workers tested the antimicrobial ability of phenolic acids
from grape seed extract on Listeria monocytogenes, Salmonella enterica serovar
typhimurium, Staphylococcus aureus, Escherichia coli, Enterococcus faecium,
Enterococcus faecalis, and Brochothrix thermosphacta, and the result showed a
inhibition of all the Gram-positive bacteria tested.43 This result corroborates many
previous reports and mechanisms of Gram-positive inhibition are proposed in
Corrales’s article.43
1.4.2 Natural chemicals from Pteridium aquilinum
1.4.2.1 Ptaquiloside
Ptaquiloside (C20H30O8) is a colourless amorphous compound which can be easily
extracted by water from every part of bracken fern. Since it exhibits strong toxicity
and water solubility, much research has been undertaken to analyse the soil
contamination caused by rain water flushed ptaquiloside. In 1983, the isolation of
ptaquiloside was first achieved by Yamada and co-workers, two years later, they
modified the isolation method and achieved a higher yield and a simpler procedure.6
OH
O
H
HO
O
O
OH
OH
OH
Figure 1.11: Chemical structure of ptaquiloside
35
Chapter 1
Introduction
Ptaquiloside is readily soluble in water and easily decomposes in aqueous solution
depending on the pH.6 In acidic conditions, ptaquiloside gradually converts to
pterosin B through acidic hydrolysis. Meanwhile, in weak alkaline conditions,
ptaquiloside will hydrolyse to a very unstable conjugated dienone named bracken
dienone, which will itself rapidly convert to pterosin B in weak acidic conditions as
shown in figure 1.12.
OH
O
H
HO
O
Ptaquiloside
O
OH
OH
OH
D-glucose
D-glucose
+
-
H /H2O
OH
(pH 8-11)
OH
O
O
HO
+
H /H2O
Pterosin B
Figure 1.12: Ptaquiloside reactions
The toxicity and carcinogenicity of bracken fern were first reported in 1960 and
Ptaquiloside was confirmed as the main carcinogen in Pteridium aquilinum in 1984:
the peritoneal injection done by Yamato group to rats induced mammary cancer by
100% and ileal cancer by 91%.6 Hirono’s group also showed that administration of
ptaquiloside via an oral route induced thrombocytopenia, myeloid aplasia and
36
Chapter 1
Introduction
neutropenia in a calf typical of the acute haemorrhagic syndrome.44 Possible
mechanism of action of ptaquiloside and related cancer model are also well discussed
in previous reports.44
1.4.2.2 Pterosins
The pterosins are a large group of sesquiterpenes and norsesquiterpenes first isolated
from Pteridium aquilinum in Japan. These compounds have been found to occur
widely in various fern species and they are also present in certain fungi of the class
basidiomycetes.45 More than one pterosin will be present after hydrolysis of the
unstable precursor ptaquiloside. Castillo and co-workers isolated pterosin A, B, K, Z
at the same time after treatment of ptaquiloside with base then acid, the HPLC and
NMR data are all well recorded.46
CH3
CH3
O
HO
HO
CH3
CH3
CH2OH
H3C
H3C
Pterosin A
CH3
Cl
Pterosin B
CH3
O
HO
CH3
CH2OH
H3C
O
CH3
CH3
H3C
Pterosin K
Pterosin Z
Figure 1.13: Chemical structures
of some bracken pterosins
37
O
Chapter 1
Introduction
During the last decade many reports based on biological activities of pterosins have
been published. When Natoris group isolated ptaquiloside and related pterosins and
tested their carcigenicity respectively, none of the pterosin was found to be
carcinogenic although some of them were cytotoxic to HeLa cells.7 The report of
Kobayashi and co-workers showed that several pterosins produce cytotoxic effects on
the ciliate, Paramecium caudatum, and also induced abnormal development of the
urchin embryos.47 Pterosin O was isolated from Pteris inequalis, and was showed to
be active against Bacillus subtilis.48 Pterosin C, which has been isolated from a
plateau plant Pteris wallichiana which grows in the Himalaya and was reported to be
active against Staphylococcus aureus.49 Pterosin Z has been shown to be a smooth
muscle relaxant,9 and has also been used in Guyana for treatment of venereal
diseases.49
Recently in 2010, pterosins are discovered to have anti-diabetic and anti-obesity
activities.50 Totally 84 different pterosins or pterosin related compounds were
synthesized or extracted from plants and have been tested in vitro and in vivo for their
anti-diabetic and anti-obesity activities. Fifteen of them expressed very strong
biological activities against type 1 and type 2 diabetes and obesity and among these
15 pterosins, four of them are naturally occurring in bracken fern and other plant
materials. These pterosin compounds were indicated to significantly lowered blood
glucose levels in STZ-induced diabetic mice, and also dramatically enhanced the
insulin sensitivity and glucose consumption in vitro. It was suggested that the pterosin
compounds may activate AMPK (AMP-activated protein kinase), which in turn
regulates insulin control of carbohydrate and fatty acid metabolism, therefore these
compounds can be considered as a potential anti-diabetic and anti-obesity agents.
Further results also indicated that these pterosin compounds can significantly reduce
serum lipids and exhibited anti-obesity effects in high fat-diet fed mice.
38
Chapter 1
Introduction
Most pterosin compounds are synthesized through a series of complex processes. For
example, McMorris and co-workers reported that pterosin F can be produced by
mixing pterosin B and triphenylphosphine in carbon tetrachloride and refluxing for 3
hours.51 The starting material pterosin B was obtained from pterosin O by
demethylation with 48% hydrobromic acid in acetic acid. Moreover, pterosin O was
afforded through catalytic hydrogenation with 5% Pd on carbon in ethyl acetate from
an intermediate indandione, and this key intermediate was formed through a
Friedel-Crafts bisacylation of the methyl ether of 2-(2,6-dimethylpheny1)ethanol with
methylmalonyl chloride. The synthesis processes are very complicated and will
generate large amounts of chemical waste and solid waste. Therefore, if naturally
occurring valuable pterosins can be extracted from bracken by using environmental
benign technologies, it will not only add value to this troublesome natural plant, but
also make some contribution to the green chemistry application in this research field.
1.5 Extraction and analysis methods
1.5.1 Supercritical Carbon Dioxide Extraction
Supercritical fluids extraction (SFE) is increasingly replacing the organic solvents that
are used in sample extraction and preparation prior to the analysis of compounds in
natural product matrices and supercritical fluid extraction plants are operating at
throughputs of up to 40,000 tons/yr in the food industry.52 The extractions of caffeine
from coffee and bitter acids from hops are well known processes performed on an
industrial scale. Coffee and tea are decaffeinated via supercritical fluid extraction and
most global brewers use extracts that are prepared from hops using supercritical fluids.
At the same time, many other applications of supercritical fluids have been
investigated. Some of these include the refining of triglycerides and fatty acids, the
production of flavours, spices and essential oil extracts from a variety of natural
39
Chapter 1
Introduction
products, the production of antioxidants, the production of low fat and low cholesterol
foods and the selective separation of nicotine from tobacco.52 Moreover, SFE
processes are also being commercialized in the polymer, pharmaceutical, specialty
lubricants and fine chemicals industries.53 To conclude, SFE are advantageously
applied to increase product performance to levels that cannot be achieved by
traditional processing technologies, and such applications for SCF offer the potential
for both technical and economic benefit.
Table 1.4: Comparison of physical and transport properties of gases, liquids, and SCF54
Property
Density(kg/m3)
Viscosity(cP)
Diffusivity(mm2/s)
Gas
1
0.01
1-10
SCF
100-800
0.05-0.1
0.01-0.1
Liquid
1000
0.5-1.0
0.001
A supercritical fluid is characterized by physical and thermal properties that are
between those of the pure liquid and gas. The fluid density is a strong function of the
temperature and pressure. The diffusivity of supercritical fluid is much higher than
that for a liquid and SCF can readily penetrates porous and fibrous solids.
Figure 1.14: Phase diagram for supercritical CO2
40
Chapter 1
Introduction
310K
320K
330K
Figure 1.15: Variation of density with pressure for carbon dioxide55
41
Chapter 1
Introduction
Carbon dioxide is a perfect selective extraction solvent since it is non-toxic,
environmentally safe and cheap. Carbon dioxide has the advantage over liquid
solvents that it’s selectivity or solvent power is adjustable and can be set to values
ranging from gas to liquid-like by changing pressure and by adding co-solvents,
among other variables.56
Supercritical CO2 is capable of extracting a wide range of diverse compounds from a
variety of sample matrices. Many non-polar to moderately polar compounds from
triglycerides to steroids can be extracted with pure carbon dioxide, whereas more
polar compounds such as phenolics, glycosidic compounds and alkaloids can be
extracted with carbon dioxide modified with entrainer solvents. Adding polar
co-solvents to the supercritical CO2 is the most common way to greatly increase the
solubility of polar components. Frequently used co-solvents include methanol, ethanol,
acetonitrile, acetone, water, diethyl ether and dichloromethane. Methanol is the most
commonly used because it is an effective polar modifier and is up to 20% miscible
with CO2. However, ethanol may be a better choice in SFE of nutraceuticals
and
food ingredients because of its lower toxicity.57 Some research has indicated that
ethanol and methanol improve the solubility of scCO2 to triterpenoids and sterols by
up to two orders of magnitude at a ethanol concentration of several mole percent.52
Figure 1.16: A detailed diagram of supercritical CO2 extraction apparatus72
42
Chapter 1
Introduction
Table 1.5: Co-solvents used to modify carbon dioxide for SFE52
o
Molecular
Pc (atm)
mass
Dielectric
constant
at 20oC
Polarity
index
Modifier
Tc ( C)
Methanol
239.4
79.9
32.04
32.70
5.1
Ethanol
243.0
63.0
46.07
24.3
4.3
Propan-1-ol
263.5
51.0
60.10
20.33
4.0
Propan-2-ol
235.1
47.0
60.10
19.3
3.9
Hexan-1-ol
336.8
40.0
102.18
13.3
3.5
2-Methoxyethanol
302
52.2
76.10
16.93
5.5
THF
267.0
51.2
72.11
7.58
4.0
1,4-Dioxane
314
51.4
88.11
2.25
4.8
Acetonitrile
275
47.7
41.05
37.5
4.8
DCM
237
60.0
84.93
8.93
3.1
Chloroform
263.2
54.2
119.38
4.81
4.1
Propylene
carbonate
352.0
---
102.09
69.0
6.1
N,N-Dimethylacet
amide
384
---
87.12
37.78
6.5
DMSO
465.0
---
78.13
46.68
7.2
Formic acid
307
---
46.02
58.5
---
Water
374.1
217.6
18.01
80.1
10.2
Carbon disulphide
279
78.0
76.13
2.64
---
43
Chapter 1
Introduction
Compared to supercritical carbon dioxide, liquid CO2 has its own special advantages.
Liquid CO2 extractions are desirable for high value thermolabile product such as plant
essential oils that require mild extraction conditions. It was reported by Moyler that
cleaner essential oil extract can be obtained by liquid CO2 which resembled much
closer the natural aroma of the original vegetable matrix compared to extraction with
supercritical CO2.4
Table 1.6 shows the selected solubility of some organic compounds in liquid CO2.
Some features of liquid CO2 can be observed from this data. Generally, liquid CO2 is
a good solvent for aliphatic hydrocarbons and most small aromatic hydrocarbons. It
behaves mostly like a hydrocarbon solvent, but with some points of difference (for
instance methanol miscibility). Halohydrocarbon, aldehydes, esters, ketones and low
carbon alcohols are readily soluble in liquid CO2. Phenols, anilines, hydroquinone and
other polyhydroxy aromatics are mainly insoluble. Polar compounds like amide, ureas
and urethanes also have very low solubility. Ordinarily, liquid CO2 can only dissolve
compounds with a molecular weight under 500 of any structural type.5
44
Chapter 1
Introduction
Table 1.6: Selected solubility of organic compounds in liquid CO25
Compound
Solubility, wt%
n-heptane
Miscible
n-dodecane
Miscible
n-hexadecane
8
n-tetracosane
1-2
-carotene
0.01-0.05
p-xylene
4-25
Pentamethylbenzene
17
Biphenyl
2
Anthracene
＜0.02
Benzotrichloride
2
Methanol
Miscible
tert-butyl alcohol
Miscible
7-tridecanol
11
p-benzoquinone
7
Benzophenone
4
Cholestanone
1.5
Methyl benzoate
Miscible
Diethyl phthalate
10
n-butyl hexadecanoate
3
N,N-dimethylaniline
Miscible
Aniline
3
Diphenylamine
1
Phenol
3
p-isopropylphenol
6
Hydroquinone
＜0.01
4-hydroxybiphenyl
0.05
a-tocopherol
1
Acetic acid
Miscible
Phenylacetic acid
＜0.1
Lauric acid
1
2,4-dinitrotoluene
24
2,4-dinitrochlorobenzene
11
Dychlorohexyl-18-crown-6
1
glucose
0
45
Chapter 1
Introduction
To conclude, extraction with liquid or supercritical carbon dioxide is an outstanding
environmentally benign extraction system with the following advantages:
 CO2 is non-toxic, odourless and leaves no solvent residues making it ideal for
food or pharmaceutical applications.
 CO2 is tuneable across a wide range of temperatures and pressures and can
selectively extract groups of molecules. In addition, with the help of co-solvents,
the solubility to polar compounds in CO2 can be extended significantly. The use
of a green co-solvent such as bio-ethanol will make scCO2 extraction of more
profound significance.
 CO2 provides an inert and neutral environment for the extraction of many
thermolabile compounds at low temperatures and provides cleaner plant extracts
because the degradation of certain compounds by lengthy exposure to high
temperatures, low pH or oxygen is avoided
 Chlorophylls are insoluble in scCO2.58 This is important in many areas of natural
product extractions and pharmaceutical processes.
 SFE extracts are also more concentrated.58
1.5.2 Soxhlet extraction
Soxhlet extraction is a very efficient method to extract compounds from solid
materials by using a hot, pure solvent. Soxhlet extraction is a general and well
established technique that outperforms other conventional extraction techniques
except for, in a limited field of applications, the extraction of thermolabile
compounds.59
In a typical Soxhlet extraction a tube of ground plant material (for increasing
solid-liquid contact area) is placed in a thimble-holder, and filled with condensed
fresh solvent from a round-bottom flask. When the liquid reaches the overflow level, a
46
Chapter 1
Introduction
siphon aspirates the extract liquid and drains it back into the round-bottom flask. In
the solvent flask, solute is separated from the solvent using distillation. By repeating
this, solid material can be extracted continuously by pure solvent and the extract will
be enriched in the round-bottom flask. The operation is repeated until complete
extraction is achieved.
1: Stirrer bar or anti-bumping granules
2: Still pot (round bottom flask) -still
pot should not be overfilled and the
volume of solvent in the still pot should
be 3 to 4 times the volume of the
soxhlet chamber.
3: Distillation path
4: Soxhlet Thimble
5: Extraction solid (residue solid)
6: Syphon arm inlet
7: Syphon arm outlet
8: Expansion adapter
9: Condenser
10: Cooling water in
11: Cooling water out
Figure 1.17: A schematic diagram of soxhlet
extraction device
Choosing a suitable solvent for soxhlet extraction is essential. Different solvents will
yield different extracts and extract compositions.60 The most widely-used solvent to
extract edible oils from plant sources is hexane. It is an excellent oil solvent due to its
oil solubility and ease of recovery. However, n-hexane, the main component of
commercial hexane, is listed as No. 1 on the list of 189 hazardous air pollutants by the
US Environmental Protection Agency.61 The use of alternative solvents such as
47
Chapter 1
Introduction
bio-ethanol, hydrocarbons, and even water, has been investigated due to
environmental, health, and safety considerations. It has been described that
supercritical carbon dioxide has extraction ability similar to that of hexane since it has
been proved to have low polarity which sits between that of hexane and toluene.62
Since hexane and toluene have been demonstrated to be excellent solvents for
extracting lipid, supercritical CO2 also has high potential to be good solvent for
extraction of triterpenoids and plant lipids. Entrainers are sometimes added into
supercritical fluid order to increase the polarity of the liquid phase. A mixture of
solvents such as isopropanol and hexane has been reported to increase the yield and
improve the kinetics of extraction.57, 63
Table 1.7: The advantages and disadvantages of conventional Soxhlet extraction57
Advantages
Disadvantages
1
Transfer equilibrium is obtained by
recycling solvent repeatedly
1
The extraction time is long
2
Maintains a relatively high
2
A large amount of solvent is used
which requires an evaporation
procedure
Agitation cannot be provided in
the Soxhlet device to accelerate
extraction rate
Thermolaible components can’t be
extracted by soxhlet
extraction temperature
3
No filtration requirement after
3
leaching
4
The Soxhlet method is simple and
4
cheap
1.5.3 Supercritical fluid chromatography
Supercritical fluid chromatography (SFC) is a relative new chromatographic
technique which has been commercially used since only the end of the last century.
The principles of operation of SFC are similar to those of HPLC, though SFC requires
a supercritical fluid (typically carbon dioxide) with an organic modifier as mobile
phase and operates in normal phase rather than reverse phase. Because of the unique
48
Chapter 1
Introduction
benefit of supercritical fluid, SFC has some significant advantages over other
separation techniques such as gas chromatography (GC) and high performance liquid
chromatography (HPLC). SFC is able to give a good separation and determination of
group of compounds that are not conveniently handled by either GC or LC. It can
analyse and purify compounds with high boiling point and low volatility that cannot
be analysed by GC and can achieve higher analysing speed and better resolution
compared with HPLC. For instance, GC is not able to analyse any non-volatile or
thermally unstable compounds. Similarly, LC cannot be used for compounds with
some functional groups that cannot be detected by either spectroscopic or using
electrochemical detectors. However compared with these conventional techniques,
SFC is relatively recent and there is a large amount of research currently underway
both in SFC method development and in hardware development.
ABPR
PDA detector
CO2 and co-solvent pump
Fraction collector
Auto
sample
Column oven
Figure 1.18: Semi-prep scale supercritical fluid
chromatography in Green Chemistry group,
University of York
49
Chapter 1
Introduction
There are many obvious advantages of using SFC. As the diffusion coefficient of
supercritical fluid is much greater than a liquid, SFC can achieve better resolution and
faster analysis speed at same condition than using HPLC with a liquid mobile phase.
Viscosity is also a major reason that SFC is superior to HPLC. Since the viscosity of
supercritical fluid is similar to that of gas, the pressure drop across the column will be
much lower than a HPLC column in operation. The low viscosity and low pressure
drop also allow higher flow rate in SFC. Generally, the optimal flow rate of SFC is
3-5 times higher than a HPLC. Thus, higher throughput can be achieved and many
more samples would be analysed within a certain period of time. Due to the unique
properties of supercritical fluid, re-equilibration could be much faster than HPLC as
well so cycle time for each sample will be greatly reduced. Generally speaking, SFC
is able to accelerate every step in sample analysis and massive analysing time will be
saved by using SFC.
SFC columns are normally packed under 1000 bar, so compared with HPLC columns
they are more robust and trustworthy to give out reliable data. It is also proven that
SFC is better for separating isomers, enantiomers and structurally related compounds.
Besides, SFC is particularly useful for the analysis of polar compounds that have less
selectivity on reverse phase LC.64
Figure 1.19: Polarity comparison between modified CO2
with common LC solvents
50
Chapter 1
Introduction
Normally, different solvents are used for HPLC due to their various polarities.
Common solvents used are hexane, DCM and isopropanol for normal phase and
acetonitrile, methanol and water for reverse phase. With the help of organic modifier,
supercritical CO2 can reach a broad range of polarity from hexane up to methanol.
The use of SFC reduces solvent volumes and therefore cost, the only thing that
modified CO2 cannot compete with is water. Overall, supercritical fluid
chromatography has shown superior performance in analysing compounds that are
very difficult to measure and separate by conventional GC and LC, and can also make
the whole analysing process easier and quicker. SFC is an ideal complement to GC
and HPLC, but new applications and development are needed to consolidate this
technique.
51
Chapter 1
Introduction
52
Chapter 2
Supercritical carbon dioxide extractions of heather
Supercritical carbon dioxide
extractions of heather
Chapter 2
53
Chapter 2
Supercritical carbon dioxide extractions of heather
2.1 Supercritical carbon dioxide extractions of heather
Triterpenoids have been reported to be present in heather at a quite high concentration;
however they are not essential for plant development.2 Prior studies indicated that the
triterpenoids within Calluna vulgaris were found to be 12-unsaturated triterpenoids
with the double bond positioned on the 12 carbon.2 These unsaturated triterpenoids
included -amyrin, -amyrin, oleanolic acid and ursolic acid.
20
19
12
C
3
8
B
17
16
15
7
5
4
D
14
9
10
22
13
1
A
E
18
11
2
21
6
Figure 2.1: Numbering of carbon skeleton of triterpenoids
Pancost and co-workers quantified the content of triterpenoids in Calluna vulgaris
and another heather species Erica tetralix.65 Their results indicated that Calluna
vulgaris contain much more triterpenoids in stems and leaves than Erica tetralix, but
less in roots. Table 2.1 shows the abundances of triterpenoids in these two heather
species.
54
Chapter 2
Supercritical carbon dioxide extractions of heather
Table 2.1 Abundances (in g/g dry plant) of triterpenoids in leaves, stems and roots of
Calluna vulgaris and Erica tetralix65
Erica tetralix
Calluna vulgaris
Coarse
Stems and
Fine roots
Stems
Leaves
Roots
roots
leaves
-amyrin
3100
1800
620
35000
370
3900
-amyrin
910
710
1800
13000
550
5700
Lupeol
780
600
120
19000
0
0
Oleanolic acid
1800
1000
600
11000
55
14000
Taraxerol
0
0
0
0
610
0
Taraxer-4-one
0
0
0
0
490
0
Ursolic acid
12000
2900
2400
52000
130
34000
1700
1300
600
26000
20
7500
Unsaturated
ursolic acid
Among these triterpenoids, ursolic acid is present in the greatest concentration in
Calluna vulgaris with -amyrin, -amyrin and oleanolic acid at lower concentrations.
Previous results indicated that ursolic acid can make up 2.5% of the dried plant of
heather in flowering season.31 Ursolic acid and oleanolic acid have been shown to
have anticancer properties against leukaemia cells, gastric tumours and breast cancer
cells in vitro.66 Other biological activities of triterpenoids such as anti-inflammation,
anti-ulcer, antimicrobial and liver protection against chemical damage has also been
demonstrated.31
Since triterpenoids are found to be more concentrated in stems and leaves in Calluna
vulgaris, aerial parts of heather were harvested in October 2007, January 2008, March
2008 and August 2008 to be representative of heather from different seasons. All the
55
Chapter 2
Supercritical carbon dioxide extractions of heather
heather samples are harvested from Hole of Horcum, North York Moors National
Park (Grid reference SE 84558 93633). All the plant samples were dried and milled to
a particle size of ≤2 mm and then stored in the fridge at 2 oC prior to extraction.
Heather supercritical carbon dioxide extracts are reported to have high proportion of
triterpenoids.2 Depending on the extraction conditions, supercritical CO2 extractions
can yield higher total triterpenoids than corresponding hexane Soxhlet extractions.
Therefore, in this study, heather samples were extracted by supercritical CO2 with and
without the presence of ethanol entrainer. Extraction was carried out under
supercritical conditions at a pressure of 35 MPa, a temperature of 50 oC and a CO2
flow rate of 40 g/min, for 4 hours. When extraction was carried out with the addition
of ethanol as entrainer the ethanol flow rate was set to 4 ml/min (10% of CO2 flow).
Several triterpenoids have been discovered in these heather extracts, and some other
triterpenoids such as taraxerol which have not been previously found in Calluna
vulgaris stems and leaves are also found.
2.2 General yield of supercritical carbon dioxide extractions
The yield of heather supercritical CO2 and supercritical CO2 with ethanol entrainer
extractions are recorded. The seasonal variation in extract yield in both dry plant and
fresh plant can be seen from figure 2.2 and 2.3. It is clear that summer flowering
sample harvested in August 2008 exhibited the highest crude yield in both dry plant
and fresh plant and January winter heather, by contrast, showed the lowest yield in
dry plant and fresh plant. The yields from heather supercritical CO2 extraction with
ethanol entrainer are generally higher than that of supercritical CO2 only extraction by
approximately 0.8%-1% in dry plant and 0.7%-1% in fresh plant. However, the
seasonal variation trend in crude yield of these supercritical extractions are exactly the
same as the lowest point found in winter sample and the peak reached in summer.
56
Chapter 2
Supercritical carbon dioxide extractions of heather
4.50%
4.50%
3.90%
4.00%
Yield (%)
3.50%
3.00%
3.30%
2.80%
3.50%
2.70%
2.30%
2.50%
1.90%
2.00%
1.50%
1.00%
0.50%
0.00%
10/2007
01/2008
03/2008
08/2008
Month
Yields of supercritical carbon dioxide extractions (dry plant)
Yields of carbon dioxide+EtOH extractions (dry plant)
Figure 2.2: Extraction yield from dried heather using supercritical
CO2 and supercritical CO2 with ethanol entrainer
3.79%
4.00%
3.49%
3.50%
Yield (%)
3.00%
2.79%
2.51%
2.95%
2.33%
2.50%
1.94%
1.64%
2.00%
1.50%
1.00%
0.50%
0.00%
10/2007
01/2008
03/2008
08/2008
Month
Yields of supercritical carbon dioxide extractions (fresh plant)
Yields of carbon dioxide+EtOH extractions(fresh plant)
Figure 2.3: Extraction yield from fresh heather using
supercritical CO2 and supercritical CO2 with ethanol entrainer
57
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3 Column fractionation of heather supercritical CO2 and
supercritical CO2 with ethanol entrainer extracts and identification
of triterpenoids
Column chromatography was used to fractionate summer heather supercritical CO2
and supercritical CO2 with entrainer extracts in conjunction with thin layer
chromatography to help purify and isolate fractions rich in triterpenoids. Several
triterpenoids were discovered in flowering heather supercritical dioxide extracts.
2.3.1 Fractionation of heather supercritical CO2 extract
The method used for column chromatography and thin layer chromatography are
stated in chapter 7.3.3 and 7.3.4. 56 initial fractions were collected from the column
and they are combined based on their spot positions revealed by TLC (figure 2.4).
Eventually, 14 combined fractions were obtained, and the weight of each fraction was
recorded after evaporating the solvent using a rotary evaporator. GC-MS samples
were then made up in dichloromethane for each fraction at a concentration of 20
mg/ml.
58
Chapter 2
Supercritical carbon dioxide extractions of heather
95% hexane/5% diethyl ether
90% hexane/10% diethyl ether
Fraction
2
75% hexane/25% diethyl ether
Fraction
3
Fraction
4
Fraction
5
Fraction
6
Fraction
7
Fraction
8
Fraction
9
Figure 2.4: TLC plate with separated spots revealed from the column of supercritical CO2 extract
59
Fraction
10
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3.1.1 Molecular ion adjustment in GC-MS analysis
It is noticed that throughout the GC-MS analysis of this work, all compounds with a
molecular mass over 400 showed their molecular ion as [M+1]+• rather than M+•.
Although this phenomenon has been observed in the analysis process, the reason for it
is not understood clearly. It is likely that the high mass calibration of the Perkin Elmer
Clarus 560S mass spectrometer used within this study was malfunctioning therefore
the molecular ion for compounds which have over 400 molecular mass were not
measured very precisely. To prove this phenomenon, several standards which
represent different groups of compounds with higher than 400 molecular mass were
analysed and their mass spectrum were compared with a standard spectrum obtained
from NIST library. Squalene (MW 410), -amyrin (MW 426) and stigmasterol (MW
412) all displayed molecular ion as [M+1]+• rather than M+•. Figure 2.5, 2.6 and 2.7
show the mass spectra obtained from Perkin Elmer mass spectrometer compared with
Nist library standards.
From the spectrum of injected standards, it is clear that due to a unique instrumental
conditions, all parent ions were displayed as [M+1]+•, not M+•. Therefore in later
GC-MS analysis and compound identification process, all the [M+1]+• parent ions will
be regarded as normal M+• parent ion.
60
Chapter 2
Supercritical carbon dioxide extractions of heather
Perkin Elmer spectrum
[M+1]+•
NIST library spectrum
M+•
Figure 2.5: Mass spectrum of squalene obtained from Perkin Elmer Clarus 560S mass spectrometer compared with NIST library standard
spectrum
61
Chapter 2
Supercritical carbon dioxide extractions of heather
Perkin Elmer spectrum
[M+1]+•
218
100
OH
NIST library spectrum
M+•
50
0
27
43
20
40
(mainlib) πAmyrin
55
60
69 81
80
95
189
109 122 135 147
161 175
100
120
140
160
180
203
426
231 242
200
220
240
257 272 283 297
260
280
300
339
320
340
365
360
393
380
400
411
420
440
Figure 2.6: Mass spectrum of -amyrin obtained from Perkin Elmer Clarus 560S mass spectrometer compared with NIST library
62 spectrum
standard
Chapter 2
Supercritical carbon dioxide extractions of heather
Perkin Elmer spectrum
[M+1]+•
55
100
NIST library spectrum
69
50
83
43
M+•
91
133
105
145 159
119
255
173
29
0
185 199
15
20
40
(mainlib) Stigmasterol
60
80
100
120
140
160
180
200
412
HO
271
300
213
229 241
220
240
285
260
280
314
300
351
327 339
320
340
369
360
397
380
400
420
Figure 2.7: Mass spectrum of stigmasterol obtained from Perkin Elmer Clarus 560S mass spectrometer compared with NIST library standard
spectrum
63
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3.1.2 Identification of alkanes in heather supercritical CO2 extract hexane fraction
The hexane eluent was collected from the column and combined as one fraction (25.2
mg) and analysed by GC (figure 2.8).
Most of the peaks in this fraction are identified as primary alkanes as they all have
identical spectra to hydrocarbon standards. These n-alkanes within Calluna vulgaris
exhibited chain lengths in the range of C21 to C35. The major peaks appeared at
31.17 min and 32.72 min were identified as hentriacontane and tritriacontane after
comparison of their retention times with C12-C60 hydrocarbon standards. In this
fraction, it is also clear that odd numbered alkanes generally show greater peak areas
than even number alkanes. This finding is consistent with previous results and it is
suggested that the production of these compounds in plants is through a
elongation-decarboxylation mechanism of corresponding even carbon numbered fatty
acids to yield odd carbon chain length alkanes.67
64
Chapter 2
Supercritical carbon dioxide extractions of heather
C31 alkane
C33
C29
squalene
C30
C32
C21
C20
C16
C1
8
C22
C24
C26 C28
C30
C32
C12
C40
C1
4
C36
C44
Figure 2.8: Gas chromatogram of fraction 1 compared with C12-50 hydrocarbon standards
65
C50
Chapter 2
Supercritical carbon dioxide extractions of heather
Squalene was also identified in this fraction at 28.94 min. Squalene occurs naturally
in plants as a precursor in the biosynthesis of sterols.68 When campesterol and
sitosterol are synthesized in the endoplasmic reticulum via the mevalonate pathway,
production of the precursor squalene would take place at the same time.69 Figure 2.9
shows the mass spectrum of the peak at 28.94 min and the standard spectrum of
squalene obtained from the NIST library.
66
Chapter 2
Supercritical carbon dioxide extractions of heather
Figure 2.9: Mass spectrum of peak at 28.94min and standard
spectrum of squalene obtained from NIST library
67
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3.1.3 Quantification of triterpenoid
The quantification of each identified triterpenoid compound was carried out with the
application of oleanolic acid as a standard and assuming that the response factor for
structurally similar triterpenoids would be similar. Standard solutions of oleanolic
acid at different concentrations were made up and analysed by GC. The absolute areas
were recorded and a standard curve was determined. Figure 2.10 shows the standard
curve of oleanolic acid and the regressive equation. A correlative coefficient of
Area of peak (pA*s)
Millions
0.9967 was obtained from this curve.
60
y = 1E+08x - 253464
R² = 0.9967
50
40
30
20
10
0
0
0.1
0.2
0.3
0.4
0.5
0.6
-10
Concentrations of oleanolic acid (mg/ml)
Figure 2.10: The standard curve of oleanolic acid
for heather triterpenoid compound quantification
2.3.1.4 Identification of triterpenoids in heather supercritical CO2 extract fractions
Triterpenoid compounds are mainly found in fractions 2 to fraction 5 which eluted
from the silica column by hexane/diethyl ether mixture in different ratios. The weight
and triterpenoid components found in each fraction are listed in table 2.2
68
Chapter 2
Supercritical carbon dioxide extractions of heather
Table 2.2 Triterpenoid compounds found in heather supercritical CO2 extract fractions
Weight
Fraction 2
Fraction 3
83.2mg
12.3mg
Triterpenoids identified
Concentration of
each triterpenoid
(dry plant)
Taraxerone
Friedelin
13,27-cycloursane
D: C-Friedoolean-8-en-3-one
326 g/g
234g/g
91g/g
77 g/g
Taraxerol
Friedelin
Friedelinol
13,27-cycloursane
9-methyl-19-Norlanosta-5,24-dien-3-ol
-amyrin
-amyrin
81g/g
155g/g
362g/g
108g/g
80g/g
131g/g
46 g/g
Fraction 4
45.7mg
-amyrin
-amyrin
g/g
221g/g
Fraction 5
29.3mg
-amyrin
-amyrin
g/g
25g/g
Figures 2.11 and 2.12 show the gas chromatogram of fractions 2-5 with triterpenoid
group highlighted. Quantification of each identified triterpenoid compound was also
accomplished by using oleanolic acid as a standard. Figure 2.13 showed the expanded
gas chromatogram of fraction 2 with each identified triterpenoid compound
highlighted.
69
Chapter 2
Supercritical carbon dioxide extractions of heather
Fraction 2
Triterpenoid
Fraction 3
Triterpenoid
Figure 2.11: Gas chromatogram of fraction 2 and fraction 3
with triterpenoid group highlighted
70
Chapter 2
Supercritical carbon dioxide extractions of heather
Fraction 4
Triterpenoid
Fraction 5
Triterpenoid
Fraction 6
Figure 2.12: Gas chromatogram of fractions 4, 5, and 6 with triterpenoid group highlighted
71
Chapter 2
Supercritical carbon dioxide extractions of heather
, 16-Apr-2010 + 15:26:28
Taraxerone
lzhscco2-2
Scan EI+
TIC
3.72e8
32.96
100
Friedelin
34.64
35.19
%
13,27-cycloursan-3-one
D:C-Friedoolean-8-en-3-one
33.28
34.24
33.46
33.75
33.90
33.12
33.38
33.84
34.46
34.04
34.36
32.69
1
Time
32.75
32.95
33.15
33.35
33.55
33.75
33.95
34.15
34.35
34.55
34.75
34.95
35.15
35.35
35.55
35.75
Figure 2.13: Expanded gas chromatogram of fraction 2 with each identified triterpenoid compound highlighted
72
35.95
Chapter 2
Supercritical carbon dioxide extractions of heather
Taraxerone (C30H48O, also called D-Friedoolean-14-en-3-one, MW 424) was found in
fraction 2 with a retention time of 32.96 min. Taraxerone exhibits a fragmentation
pattern of m/z 424, 409, 300, 285 204 and 133. Figure 2.15 shows the mass spectrum
comparison of the peak at 32.96 min with taraxerone standard obtained from NIST
library. Taraxerone was present at 326g/g in August heather dry plant.
Figure 2.14 Chemical structure of taraxerone
An important cleavage process in this compound is the retro Diels-Alder collapse of
ring D (Type 1) as shown below yielding diene fragment A gives the characteristic
ion at m/z 300; further loss of a methyl group yields ion B (m/z 285).
Type 1
+
•
A
m/z 300
•
-CH3
B
m/z 285
Scheme 2.1: Proposed mechanism for fragmentation of taraxerone (type 1)
73
Chapter 2
Supercritical carbon dioxide extractions of heather
A second important cleavage in this compound occurs in ring C involving rupture of
the 8, 14 and 11, 12 bonds (Type 2). Previous researchers found it very difficult to
explain this fragmentation process, however they postulated a mechanism in which
the molecular ion of taraxerone generates a new carbonium ion, which then leads to
the ion C by appropriate cleavage.70 Ion C (m/z 204) is the most abundant in the
spectrum of taraxerone.
The postulated mechanism of type 2 fragmentation is not entirely convincing, but as
mentioned before, this fragmentation is difficult to explain. The mechanism of the
type 2 fragmentation showed in scheme 2.2 is only a tentative mechanism based on
previous researcher’s postulation.
Type 2
12
11
14
.
8
.
.
C
m/z 204
Scheme 2.2: Proposed mechanism for fragmentation of taraxerone
(type 2)
74
Chapter 2
Supercritical carbon dioxide extractions of heather
C
A
B
NIST 2175: D-FRIEDOOLEAN-14-EN-3-ONE
(TARAXERONE)
Figure 2.15: Mass spectrum of the peak at 32.96min in
fraction 2 compared with standard taraxerone spectrum
75
Chapter 2
Supercritical carbon dioxide extractions of heather
The peak with a retention time of 32.93 min in fraction 3 was identified as taraxerol
(C30H50O, MW 426) by directly comparing its mass spectrum with standard
taraxerol spectrum. Taraxerol gives characteristic fragmentation of m/z 426, 411,
302, 287 and 204. Figure 2.17 shows the comparison between the spectrum of peak
at 32.93 min and the standard spectrum of taraxerol obtained from NIST library.
Taraxerol showed a concentration of 81 g/g in heather dry plant.
Figure 2.16 Chemical structure of taraxerol
The same retro Diels-Alder collapse of ring D (Type 1) occurs during taraxerol
fragmentation process and affords the characteristic ion A at m/z 302. A further loss
of a methyl group yields ion B (m/z 287). Similarly, rupture of the 8, 14 and 11, 12
bonds occurs in ring C and yield the characteristic ion C (m/z 204).70
76
Chapter 2
Supercritical carbon dioxide extractions of heather
Type 1
+
•
A
m/z 302
•
-CH3
B
m/z 287
Type 2
12
11
14
.
.
8
.
C
m/z 204
Scheme 2.3: Proposed mechanism for fragmentation of taraxerol
77
Chapter 2
Supercritical carbon dioxide extractions of heather
According to Pancost and co-workers,65 taraxerone and taraxerol were only detected
in the roots of Calluna vulgaris. Their results indicated that taraxerone and taraxerol
were not detected in aerial part includes stems and leaves. However, in this study,
taraxerone and taraxerol were both found in heather aerial parts, the supercritical
carbon dioxide extract showed a high concentration of taraxerone and was a major
component in this fraction.
Taraxerol and taraxerone are both reported to have biological activities such as
anti-cancer and anti-inflammatory. Taraxerone showed in vitro anti-leishmanial
activity against promastigotes of Leishmania donovani and anti-tumour activity on
K562 leukemic cell line.71 Taraxerol was reported to have significant activity against
dermatophytes T. rubrum and T. mentagrophytes with MIC values between 6.2 and
25 g/ml, additionally, taraxerol also expresses moderate activity against
Aspergillus niger.72
78
Chapter 2
Supercritical carbon dioxide extractions of heather
C
B
A
Figure 2.17: Mass spectrum of peak at 32.93min in fraction
3 compared with standard taraxerol spectrum
79
Chapter 2
Supercritical carbon dioxide extractions of heather
The peaks at 34.64 min in fraction 2 and 34.11 min in fraction 3 which expressed
quite high absolute areas were identified as friedelin (C30H50O, MW 426). Friedelin
exhibited a fragmentation pattern of m/z 411, 273, 123 and 109. The mass
fragmentation patterns of friedelin were presented in Scheme 2.3. The molecular ion
peak of this compound was observed at m/z 426. The loss of a methyl group was
indicated by the presence of a peak at m/z 411. Other significant ions at m/z 341,
273, 205 and 123 were attributed to the fragmentation of A, B, C and D rings
respectively. Figure 2.18 shows the mass spectrum of the peak at 34.63 min in
fraction 2 compared with standard spectrum of friedelin. The concentration of
friedelin in was calculated as 234 g/g in fraction 2 and 155 g/g in fraction 3.
Therefore, its total concentration in heather dry plant is 389 g/g.
Friedelin and several derivatives of it have been tested on their biological activities.
Friedelin has been tested for its cytotoxicity on human leukemia cell HL-60, human
ovarian cancer cell SK-OV-3, human lung adenocarcinoma cell A549 and human
colon cancer HT-29 cell lines.73 Friedelin showed significant cytotoxic activity
against all cell lines with GI50 values in the range of 11.1 to 13.5 M. Friedelin has
also been evaluated for its anti-inflammatory effects in non-cytotoxic concentrations
(1-100 nM).73 The result shows that friedelin expressed moderate inhibitory activity
of TNF- secretion in the presence of LPS in a murine macrophage cell line at 100
nM. Other researchers reported that the anti-inflammatory activity of friedelin can
be tripled by introducing hydroxyl group into its 3 position.74 Therefore, friedelin
not only has valuable biological activities on its own, but also has a good potential to
be applied as pharmaceutical precursor.
80
Chapter 2
Supercritical carbon dioxide extractions of heather
+
.
E
C
A
•
-CH3
D
m/z 411
B
+
E
C
A
D-ring
cleavage
D
D
B
m/z 123
+
E
C
A
C-ring
cleavage
D
C
B
m/z 205
+
E
C
A
B
D
B-ring
cleavage
B
m/z 273
81
Chapter 2
Supercritical carbon dioxide extractions of heather
+
E
C
A
D
A-ring
cleavage
A
B
m/z 341
Scheme 2.4: Proposed mechanism for fragmentation of friedelin
82
Chapter 2
Supercritical carbon dioxide extractions of heather
, 16-Apr-2010 + 15:26:28
lzhscco2-2 4738 (34.644) Cm (4735:4742-4757:4766)
69
100
67
81
83
79
%
43
57
53
D
95
55
65
71
Scan EI+
1.15e7
96
109
C
123
B
121
97
133 137
84
149
163
179 191
205
218 231
246 257
0
35
55
75
95
115
135
155
175
195
215
235
255
273
302
287
275
295
327
315
341
335
412
355
375
395
415
427
435
69
100
95
109
55
125
81
50
41
163
137
205
273
218
179
149
191
231
246
29
0
A
302
257
426
O
287
313 326
20
40
60
(mainlib) Friedelan-3-one
80
100
120
140
160
180
200
220
240
260
280
411
341
300
320
353
340
369 382 397
360
380
400
420
Figure 2.18: The mass spectrum of the peak at 34.63min in fraction 2 compared with standard spectrum of friedelin
83
440
m/z
Chapter 2
Supercritical carbon dioxide extractions of heather
13,27-cycloursan-3-one (C30H48O, MW 424) was found in both fraction 2 and
fraction 3. The peaks present at 33.28 min in fraction 2 and 33.39 min in fraction 3
have the identical mass spectrum and were identified as 13,27-cycloursan-3-one
after compared their mass spectrum with standard spectrum obtained from NIST
library. 13,27-cycloursan-3-one exhibited a characteristic fragmentation pattern as
m/z 409, 205 and 138. Figure 2.20 shows the mass spectrum of the peak at 33.28
min in fraction 2 with the standard spectrum of 13,27-cycloursan-3-one.
Figure 2.19: Chemical structure of 13,27-cycloursan-3-one
Cyclopropyl ursane-type triterpenoid are very rarely found in narual plant. It’s only
been reported once that this compound has been discovered in African tree
Phyllanthus polyanthus,75 however, it’s been synthesised by Barton & de Mayo in
1953.76 The fragmentation mechanism of 13,27-cycloursan-3-one is not fully
understood. The forming of ion at m/z 205 was speculated due to a cleavage in ring
C. Therefore, the idenfication of 13,27- cycloursan-3-one is a tentative identification
based on the direct comparison between the mass spectrum of the peak and standard
spectrum. The quantity of 13,27-cycloursan-3-one was found to be 91 g/g in
fraction 2 and 108 g/g in fraction 3. Therefore the total concentration of
13,27-cycloursan-3-one in heather supercritical CO2 extract is 199 g/g. At the
moment, no literature about any biological activity of this triterpenoid compound
has been published.
84
Chapter 2
Supercritical carbon dioxide extractions of heather
M+•
[M-CH3]+
Figure 2.20: The mass spectrum of the peak at 33.28min in fraction 2
compared with standard spectrum of 13,27-cycloursan-3-one
85
Chapter 2
Supercritical carbon dioxide extractions of heather
9-methyl-19-Norlanosta-5,24-dien-3-ol (C30H50O, MW 426) was detected in fraction
3 at 33.60 min. Due to the standard mass spectrum in NIST library is incomplete (no
ion below m/z 100 was recorded), the mass spectrum were not matched perfectly,
therefore the identification is only a tentative identification based on the direct
comparison between the mass spectrum of the peak and standard spectrum of
9-methyl-19-Norlanosta-5,24-dien-3-ol. 9-methyl-19-Norlanosta-5,24-dien-3-ol is
also called as boeticol or (+)-antiquol B, and it’s been reported to be found in
Euphorbia boetica.77 This previous research paper also recorded the NMR and
GC-MS spectrum of this compound. Boeticol yielded a series of ions as m/z 426,
408, 393, 313, 295, 274, 259, 231, 205, 163, 134, 123, 121, 107, 95, 81 and 69.
Since the research paper did not present the spectrum of boeticol, figure 2.21
indicates a proposed based on the distribution of fragments according to the author.
Figure
2.22
shows
the
incomplete
9-methyl-19-Norlanosta-5,24-dien-3-ol
from
standard
NIST;
mass
library.
spectrum
The
of
proposed
fragmentation mechanism of this compound was also discussed in scheme 2.5. The
concentration of this triterpenoid was measured as 80 g/g in dry plant.
86
Chapter 2
Supercritical carbon dioxide extractions of heather
+
•
A
m/z 274
•
-CH3
B
m/z 259
+
•
+
•
-H2O
C
m/z 408
•
-CH3
D
m/z 393
Scheme 2.5: Proposed mechanism for fragmentation of
9-methyl-19-Norlanosta-5,24-dien-3-ol
87
Chapter 2
Supercritical carbon dioxide extractions of heather
B
A
Pictorial standard spectrum of
9-methyl-19-Norlanosta-5,24-dien-3-ol
according to Jose’s 77 report
Figure 2.21: The mass spectrum of the peak at 33.60 min in fraction 3 compared with
standard spectrum of 9-methyl-19-Norlanosta-5,24-dien-3-ol in previous research77
88
D
C
Chapter 2
Supercritical carbon dioxide extractions of heather
A
426
B
100
274
259
C
134
163
123
50
109
149
205
D
231
HO
0
408
393
100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440
(mainlib) 19-Norlanosta-5,24-dien-3-ol, 9-methyl-, (3β,9β,10α)-
Figure 2.22: The tandard mass spectrum of
9-methyl-19-Norlanosta-5,24-dien-3-ol in Nist library
89
Chapter 2
Supercritical carbon dioxide extractions of heather
-amyrin and -amyrin (C30H50O, MW 426) are very important triterpenoid
compounds in heather. In previous research, -amyrin has been reported to have the
highest concentration among all identified triterpenoid compounds in heather.2 In
this study, -amyrin and -amyrin are present in fraction 3 to fraction 5 consistently
and afford the major constituent of heather total triterpenoids. The peaks at 33.52
min in fraction 3, 32.97 min in fraction 4 and 32.97 min in fraction 5 are identified
as -amyrin. The peaks at 33.17 min in fraction 3, 32.64 min in fraction 4 and 32.62
min in fraction 5 are identified as -amyrin. Since the spectrum of -amyrin and
-amyrin are very similar, Kovats index was also applied to help identify peaks
besides comparing spectrum directly. According to the standard KI value data
published by NIST Mass Spectrometry Data Centre, -amyrin has lower KI value
than -amyrin on a DB-5 column. Therefore, the peak with shorter retention time
was identified as -amyrin, and the one with longer retention time was identified as
-amyrin. -amyrin and -amyrin both have a fragmentation pattern of m/z 218 and
203.
+
•
-amyrin: R1=CH3 R2=H
-amyrin: R1=H R2=CH3
-amyrin A1: R1=CH3 R2=H
-amyrin A2: R1=H R2=CH3
m/z 218
•
-CH3
B
m/z 203
Scheme 2.6: Proposed mechanism for fragmentation of -amyrin and -amyrin
90
Chapter 2
Supercritical carbon dioxide extractions of heather
A typical retro Diels-Alder collapse of ring C happens during -amyrin and
-amyrin fragmentation and affords characteristic ions A1 and A2 at m/z 218 and
further loss of a methyl group yields ion m/z 203 (Scheme 2.6). Figure 2.23 and 2.24
show the standard spectrum of these two compounds with the spectrum of
corresponding peaks.
The concentration of -amyrin was calculated as 131g/g in fraction 3, g/g in
fraction 4 and g/g in fraction 5. That makes a total concentration of 693 g/g in
heather dry plant. The concentration of -amyrin was calculated as 46g/g in
fraction 3, 221g/g in fraction 4 and 25g/g in fraction 5. That makes a total
concentration of 292 g/g in dry heather plant.
91
Chapter 2
Supercritical carbon dioxide extractions of heather
SCCO2 extracted heather, fractionated
, 27-Apr-2010 + 12:12:40
A1
lzhscco2-4 4568 (33.519) Cm (4565:4571-4573:4574)
B
100
55
69
43
%
81
67
41
53
77
65
91
95
107
97
85
122 133
117
131
189
147
161
150
203
175 187 190
219
218
229
0
35
55
75
95
115
135
155
175
Scan EI+
7.88e6
218
195
215
235
258
243
255
273
287 297
275
295
315 327
315
366
341
335
355
394
375
395
427
412
415
m/z
435
218
100
OH
50
0
28
43 55
20
40
(mainlib) α-Amyrin
60
69 81
80
189
95 109 122 135
147 161 175
100
120
140
160
180
203
426
231 242 257 272 283 297
200
220
240
260
280
300
339
320
Figure 2.23: The mass spectrum of the peak at 33.52 min in fraction 3
compared with standard spectrum of -amyrin
92
340
365
360
393
380
400
411
420
440
Chapter 2
Supercritical carbon dioxide extractions of heather
A2
SCCO2 extracted heather, fractionated
, 27-Apr-2010 + 12:12:40
B
lzhscco2-4 4516 (33.172) Cm (4515:4518-4526:4532)
Scan EI+
4.45e6
218
100
203
55
%
43
41
69
81 91 95
67
53
77
83
107
119
97
131
135 147
189
161
175
187 191
219
204
229
62
0
35
55
75
95
115
135
155
175
195
215
244
235
257 269 281
255
275
295
313
329 341
295
315
335
355 366
355
394
375
406
395
412
427
415
m/z
435
218
100
OH
50
203
0
28
43 55
20
40
(mainlib) β-Amyrin
60
69 81
80
95 109
189
121 135 147
161 175
100
120
140
160
180
231 242 257 272 283 297
200
220
240
260
280
300
337
320
Figure 2.24: The mass spectrum of the peak at 33.17 min in fraction
3 compared with standard spectrum of -amyrin
93
340
365
360
393
380
400
411
426
420
440
Chapter 2
Supercritical carbon dioxide extractions of heather
-amyrin and -amyrin are both well researched in respect to their biological
activities. A mixture of -amyrin and -amyrin extracted from the Brazilian
medicinal herb Protium hetaphyllum was tested for its ability to inhibit aggregation
of human platelets.78 The results showed that -amyrin and -amyrin significantly
inhibited platelet aggregation by 40%, 64%, and 60% in the assay carried out with
ADP as agonist, at the doses of 100, 150, and 200 µM respectively. Several
derivatives of -amyrin and -amyrin were also reported for their anti-fungal
activities.79 -amyrin and -amyrin formate and-amyrin and -amyrin acetate
showed significant inhibition against Candida species, -amyrin and -amyrin
formate also inhibited the adhesion ability of Candida albicans to human epithelial
cells.
2.3.1.5 Composition of triterpenoid compounds in heather supercritical CO2 extract
The triterpenoid profile distribution demonstrated -amyrin is the dominant
compound within heather supercritical carbon dioxide extract. -amyrin makes up
27.74% of total triterpenoid extracted by supercritical CO2, and its concentration
was recorded as 693 g/g dry plant. Friedelin contributes the second highest total
concentration at 389g/g dry plant and this makes up 15.57% of total triterpenoid in
this extract. The total concentration of these 9 identified triterpenoid compounds is
2499g/g.
94
% Composition (Triterpenoid)
Chapter 2
Supercritical carbon dioxide extractions of heather
27.74%
30.00%
25.00%
20.00%
15.00%
15.57%
11.69%
7.97%
10.00%
5.00%
14.49%
13.05%
3.24%
3.08%
3.20%
0.00%
Figure 2.25: Triterpenoid distribution of heather
supercritical CO2 extract
2.3.2 Fractionation of heather supercritical CO2 with ethanol entrainer extract
The same methods were used for column chromatography and thin layer
chromatography to separate this extract and they are stated in chapter 7.3.3 and 7.3.4.
Totally 75 fractions were collected from the column and they are combined based on
their spot positions revealed by TLC. Eventually, 11 combined fractions were
obtained, and the weight of each fraction was recorded after evaporating the solvent
using a rotary evaporator. GC-MS samples were then made up in dichloromethane
for each fraction at a concentration of 20 mg/ml.
95
Chapter 2
Supercritical carbon dioxide extractions of heather
Fraction 1
Fraction 2
Fraction 3
100% hexane
Fraction
4
95% hexane/5% diethyl ether
Figure 2.26: TLC plate (1) with spots revealed for
supercritical CO2 with entrainer extract
96
Chapter 2
Supercritical carbon dioxide extractions of heather
90% hexane/10% diethyl ether
75% hexane/25% diethyl ether
Fraction 5
Fraction 6
Fraction 7
Fraction 8
Fraction 9
Figure 2.27: TLC plate (2) with spots revealed for
supercritical CO2 with entrainer extract
97
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3.2.1 Identification of alkanes in heather supercritical CO2 with entrainer extract
hexane fraction
The hexane eluent was collected from the column and combined as one fraction and
weight 32.8 mg. This fraction 1 was then made up a GC sample at 20 mg/ml. Gas
chromatogram of this fraction was shown in figure 2.28.
As found in fraction 1 of supercritical CO2 extract, all major peaks in this fraction
are identified as primary alkanes as they all have identical spectra with hydrocarbon
standards. The same range of chain length C21 to C35 were identified. The
dominant peaks appeared at 31.29 min and 32.81 min were identified as
hentriacontane and tritriacontane after comparison of their retention times with
C12-C60 hydrocarbon standards. Odd numbered alkanes still express greater
absolute peak areas than even number alkanes in this fraction.
98
Chapter 2
Supercritical carbon dioxide extractions of heather
Heather hexane fraction
, 22-Feb-2010 + 10:52:51
C31
lz008-f1
31.29
100
32.81
Scan EI+
TIC
9.58e8
C33
%
C29
C32
C27
C21
29.64
C30
32.06
27.89
30.48
21.81
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
Figure 2.28: Gas chromatogram of fraction 1 of heather
supercritical CO2 with ethanol entrainer extract
99
35.51
40.51
45.51
Chapter 2
Supercritical carbon dioxide extractions of heather
2.3.2.2 Identification of new triterpenoid compounds in heather supercritical CO2
with ethanol entrainer extract fractions
Triterpenoid compounds are found in fractions 2 to fraction 6 of heather
supercritical CO2 with entrainer extract. The weight and triterpenoid components
found in each fraction are listed in table 2.3. All the triterpenoid compounds
identified in this extract already present in previous supercritical CO2 extract.
Table 2.3: Triterpenoid compounds found in heather supercritical CO2 with ethanol
entrainer extract fractions
Weight
Triterpenoids identified
Concentration of
each triterpenoid
(dry plant)
36.8mg
Taraxerone
13,27-cycloursan-3-one
91 g/g
42g/g
32.4mg
Taraxerone
Friedelin
13,27-cycloursan-3-one
136g/g
157 g/g
25 g/g
Fraction 4
21.6mg
araxerol
13,27-cycloursan-3-one
D-friedooleanan-3-ol
g/g
76g/g
122 g/g
Fraction 5
37.8mg
-amyrin
-amyrin
g/g
121g/g
Fraction 6
62.6mg
-amyrin
-amyrin
g/g
47 g/g
Fraction 2
Fraction 3
2.3.2.3 Composition of triterpenoid compounds in heather supercritical CO2 with
ethanol entrainer extract
-amyrin is the dominant compound within heather supercritical CO2 with ethanol
entrainer extract as well as the supercritical extract without entrainer. However, in
this extract, the concentration of each individual triterpenoid compound seems more
100
Chapter 2
Supercritical carbon dioxide extractions of heather
even compared with supercritical CO2 extract. -amyrin composes 24.41% of total
triterpenoid in this extract, when taraxerone becomes the second greatest component
and makes up 19.04% of total triterpenoid.
The total concentration of these 7
identified triterpenoid compounds is 1192g/g.
24.41%
% Composition (triterpenoid)
25.00%
20.00%
19.04%
15.00%
12.00%
14.09%
13.17%
10.24%
10.00%
7.05%
5.00%
0.00%
Figure 2.29: Triterpenoid distribution of heather
supercritical CO2 with ethanol entrainer extract
101
Chapter 2
Supercritical carbon dioxide extractions of heather
2.4 Conclusion and further remarks
Heather samples picked in different seasons were extracted using supercritical CO2
and supercritical CO2 with ethanol entrainer. The crude extract yield from each
extraction was recorded and evaluated. It was found that summer heather picked in
August flowering season exhibited the greatest yield in both supercritical CO2 and
supercritical CO2 with entrainer extractions (3.5% and 4.5% dry plant). Therefore,
August heather extracts were selected as research object in further studies.
Column chromatography was then applied to help identify triterpenoid compound in
heather. Heather supercritical CO2 and supercritical CO2 with entrainer extracts were
fractionated with silica columns and analysed by GC-MS for their chemical
composition. Totally 9 triterpenoid compounds were identified from heather
supercritical CO2 and supercritical CO2 with entrainer extracts. Among these 9
triterpenoids, 5 of them were previously unreported in heather. Taraxerone and
taraxerol which have been proved to only be present in heather root were also
detected in these heather aerial part extracts. Quantification of each individual
triterpenoid was also accomplished by using oleanolic acid as a standard. Several
novel triterpenoid compounds found in heather have already been studied with regard
to their biological activities and some of them were proven to have good potential as
pharmaceutical drugs or drug precursors. However, a few of them seem have never
been researched before and very little literature can be found about these new
triterpenoids. Therefore, further work is required to fully study these triterpenoids for
their potential biological activities. Moreover, the identification of some triterpenoid
compounds such as 13,27-cyclousran-3-one was not conclusive as the identification
of these compounds was only tentative based on the best match of their spectrum with
corresponding standard library data. Therefore, the evidence of identification of these
compounds is not strong enough from this data alone and further supporting evidence
such as NMR data is required to give a more precise identification.
102
Chapter 2
Supercritical carbon dioxide extractions of heather
The distribution of each individual triterpenoid was also evaluated in this study. In
heather supercritical CO2 extract, -amyrin was demonstrated to be the dominant
component and made up 27.74% of total triterpenoid extracted by supercritical CO2
and has a concentration of 693 g/g in dry plant. Friedelin contributes the second
highest total concentration at 389 g/g dry plant and make up 15.57% of total
triterpenoid. The total concentration of the 9 identified triterpenoid compounds in
heather supercritical CO2 extract is 2499g/g. In heather supercritical CO2 with
ethanol entrainer extract, the distribution of each identified triterpenoid compound
seems more even compared with supercritical CO2 extract. The dominant triterpenoid
was still found to be -amyrin at a concentration of 291 g/g. It composes 24.41% of
total triterpenoid in this extract, when taraxerone becomes the second greatest
component and makes up 19.04% of total triterpenoid.
The total concentration of the
7 identified triterpenoid compounds in heather supercritical CO2 with ethanol
entrainer extract is 1192g/g. This change in distribution can be explained by the
greater solubility of the more polar triterpenoid alcohols in the supercritical CO2 with
entrainer.
Compared to the total triterpenoid result of Pancost,65 the total identified triterpenoid
content in this study is much lower. However, Pancost and co-workers extracted
Calluna vulgaris in dichloromethane and methanol mixed solvent, and they didn’t
state when their heather raw material is harvested. Andrew Hunt has done hexane
extractions and supercritical carbon dioxide extractions of heather, and his result
indicated that the concentration of heather total triterpenoid compounds is 2793 g/g
dry plant in hexane extract.2 In supercritical CO2 extracts of heather, this figure ranges
from 1652g/g to 2944 g/g dry plant depend on the extraction conditions. Therefore,
the results obtained in this study strongly corroborate with previous result, and the
total triterpenoid content of seasonal heather supercritical carbon dioxide extracts
determined by UV spectrophotometry will be discussed in detail in next chapter.
103
Chapter 2
Supercritical carbon dioxide extractions of heather
104
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
Seasonal variation of total
triterpenoids compounds in heather
Chapter 3
105
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
3.1 Seasonal variation of total triterpenoid compounds in heather
It has been reported that Calluna vulgaris contains high levels of total triterpenoid
compounds.65 Pancost and co-workers previously reported that the total triterpenoid
content can reach 65100 g/g dry plant in heather stems and leaves and Hunt2 also
indicated that the total triterpenoid content in Calluna vulgaris hexane extract is
4301g/g dry plant. Supercritical carbon dioxide extractions of heather were also
carried out by Hunt,2 and a total triterpenoid content from heather varied from
1652g/g to 2944 g/g dry plant depending on extraction conditions. However, the
seasonal variation of total triterpenoid compounds has never been reported before.
Therefore, in this study, the quantification of total triterpenoid in heather, alongside
with the seasonal variation of total triterpenoid within Calluna vulgaris were
investigated in detail.
3.2 Water content variation of heather
Water content of Calluna vulgaris was examined in samples from October 2007 to
August 2008 (Figure 3.1). The lowest water content was found to be 10.52% by total
weight in autumn heather sample (10/2007). This figure rose up to 13.75% in winter
heather (01/2008), and then the increasing trend continues until it reaches a maximum
in summer flowering heather (08/2008) at 15.72%. Spring heather (03/2008) also has
fairly high total water content at 15.47% which is only slightly lower than that of
August heather.
106
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
15.47%
16.00%
15.72%
13.75%
water content (%)
14.00%
12.00%
10.52%
10.00%
8.00%
6.00%
4.00%
2.00%
0.00%
10/2007
01/2008
03/2008
08/2008
Month
Figure 3.1: Seasonal variation in water content of Calluna
vulgaris
3.3 Total extract yield of comparison between supercritical CO2 and
hexane
Since supercritical carbon dioxide has been described to have extracting ability
similar to that of hexane,2 traditional hexane Soxhlet extractions were also applied to
give a comparison on yield and triterpenoids composition in heather. The hexane
soxhlet extractions were carried out using the identical heather samples as raw
material and were extracted for 4 hours. The crude yield comparison of heather
hexane extractions and corresponding supercritical CO2 extractions in both dry plant
and fresh plant are shown in figure 3.2 and 3.3.
107
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
4.50%
4.05%
4.00%
Yield (%)
3.50%
3.00%
3.50%
2.93%
2.80%
2.73%
2.30%
2.50%
1.90% 1.93%
2.00%
1.50%
1.00%
0.50%
0.00%
10/2007
01/2008
03/2008
08/2008
Month
Yield of supercritical carbon dioxide extractions (dry plant)
Yield of hexane soxhlet extractions (dry plant)
Figure 3.2: Comparison of hexane extractions and
supercritical extractions in dry plant
The figure shows that the yields of heather hexane soxhlet extractions are very similar
to that of corresponding supercritical CO2 extractions with differences within 0.45%
in dry plant and 0.46% in fresh plant. It is obvious that for supercritical CO2
extractions, summer flowering sample picked in August 2008 exhibited the highest
crude yield in both dry plant (3.50%) and fresh plant (2.95%) and January winter
heather, on the contrast, showed the lowest yield in dry plant (1.90%) and fresh plant
(1.64%). The seasonal yield variation of heather hexane extractions is identical to that
of supercritical CO2 extractions with the greatest yield apparent in summer and the
lowest in winter. These results also corroborate previous work with heather hexane
extract yield and seasonal variation.
108
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
3.41%
3.50%
2.95%
3.00%
2.51%
2.62%
2.31%
Yield (%)
2.50%
1.94%
2.00%
1.64% 1.66%
1.50%
1.00%
0.50%
0.00%
10/2007
01/2008
03/2008
08/2008
Month
Yield of supercritical carbon dioxide extractions (fresh plant)
Yield of hexane Soxhlet extractions (fresh plant)
Figure 3.3: Comparison of hexane extractions and
supercritical extractions in fresh plant
3.4 Seasonal variation of total triterpenoid content in heather
The quantification of total triterpenoid compounds was carried out with the
application of oleanolic acid as a standard. Vanillin-perchloric acid was applied as
colouring reagent to determine the content of total triterpenoids (method described in
7.3.7). Standard solutions containing 0, 20, 40, 60, 80, 100 and 120 g of oleanolic
acid were made up to create a calibration curve. The total triterpenoid content was
calculated as g/g dried weight of plant material using the following linear equation
based on the calibration curve:
Y=0.0049X
109
0.0111
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
where Y is the absorbance and X is weight of oleanolic acid equivalent. Figure 3.4
shows the standard curve of oleanolic acid and the regressive equation. A correlative
coefficient of 0.9981was obtained from this curve.
0.8
y = 0.0049x - 0.0111
R² = 0.9981
Absorbance at 550nm
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
20
40
60
80
100
120
140
160
-0.1
Weight of oleanolic acid (g)
Figure 3.4: The standard curve of oleanolic acid for heather
total triterpenoid quantification
Table 3.1 shows the calculated total triterpenoid content of each supercritical carbon
dioxide extract. Figure 3.5 and 3.6 shows the seasonal variation trend of total
triterpenoid content in these four heather samples in both dry plant and fresh plant.
110
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
Table 3.1: Calculated total triterpenoid content in each heather supercritical extract
Total triterpenoid in
Total triterpenoid in
supercritical CO2 extract
supercritical CO2 with ethanol
(dry plant)
entrainer extract (dry plant)
10/2007
8500 g/g
8650g/g
01/2008
4200g/g
4300g/g
03/2008
11200g/g
12000g/g
08/2008
15400g/g
16000g/g
Abundance of total triterpenoid (g/g)
Sample date
15400
16000
14000
11200
12000
10000
16000
12000
8500 8650
8000
6000
4200 4300
4000
2000
0
10/2007
01/2008
03/2008
08/2008
Month
total triterpenoid in supercritical carbon dioxide extract (dry plant)
total triterpenoid in supercritical carbon dioxide with ethanol entrainer extract (dry plant)
Figure 3.5: Total triterpenoids in supercritical CO2 and
supercritical CO2 with ethanol entrainer extracts (dry plant)
111
Abundance of total triterpenoid (g/g)
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
12979
14000
12000
9467
10000
13485
10144
7606 7740
8000
6000
3620 3707
4000
2000
0
10/2007
01/2008
03/2008
08/2008
Month
total triterpenoid in supercritical carbon dioxide extract (fresh plant)
total triterpenoid in supercritical carbon dioxide with ethanol entrainer extract (fresh plant)
Figure 3.6: Total triterpenoids in supercritical CO2 and
supercritical CO2 with ethanol entrainer extracts (fresh plant)
Generally, the total triterpenoid content in supercritical CO2 with entrainer extracts
are slightly higher than corresponding supercritical CO2 extracts. The lowest total
triterpenoid content was found in winter heather picked in January 2007, which is
4200 g/g by dry weight and 3620 g/g by fresh weight in supercritical CO2 extract,
and 4300 g/g dry plant and 3707 g/g fresh plant in supercritical CO2 with entrainer
extract. Thereafter, the total triterpenoid in heather increased dramatically and reached
11200 g/g dry plant (9467 g/g in fresh plant) in supercritical CO2 extract and 12000
g/g dry plant (10144 g/g in fresh plant) in supercritical CO2 with entrainer extract.
Then the total triterpenoid content keeps increasing until it reached the highest
amount at 15400 g/g in dry plant (12979 g/g fresh plant) in August summer heather
supercritical CO2 extract and 16000 g/g (13485 g/g fresh plant) in supercritical
CO2 with entrainer extract. This triterpenoid content makes up 1.6% of total dry
weight and 1.35% of fresh weight of summer heather. In autumn, the total triterpenoid
falls down to 8500 g/g dry plant (7606 g/g fresh plant) in supercritical CO2 extract
and 8650 g/g dry plant (7740 g/g fresh plant) in supercritical with ethanol extract.
112
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
The seasonal variation trend of total triterpenoid content is exact the same with that of
heather crude extracting yield. The greatest amount of triterpenoid content was found
in summer flowering heather and the poorest was detected in winter heather.
3.5 Comparison of total triterpenoid content extracted by
supercritical carbon dioxide and hexane
The total triterpenoid content in supercritical carbon dioxide extracts were also
compared with that of hexane extracts due to their similar extracting ability. Figure
3.7 and 3.8 showed the seasonal variation of total triterpenoid content in these two
Abundance of total triterpenoid (g/g)
series of extracts.
18000
18000
15400
16000
12400
14000
11200
12000
10000
8500
9300
8000
4200 4500
6000
4000
2000
0
10/2007
01/2008
03/2008
08/2008
Month
total triterpenoid in supercritical carbon dioxide extract (dry plant)
total triterpenoid in hexane extract (dry plant)
Figure 3.7: Total triterpenoids in supercritical CO2
and heather hexane extracts (dry plant)
113
Abundance of total triterpenoid (g/g)
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
15170
16000
12979
14000
12000
10000
10482
9467
7606
8322
8000
6000
3620
3879
4000
2000
0
10/2007
01/2008
03/2008
08/2008
Month
total triterpenoid in supercritical carbon dioxide extract (fresh plant)
total triterpenoid in hexane extract (fresh plant)
Figure 3.8: Total triterpenoids in supercritical
CO2 and heather hexane extracts (fresh plant)
From the data shown in figures 3.7 and 3.8 it is clear that the seasonal variation of
total triterpenoid in hexane extract is identical with that in supercritical CO2 extract.
The lowest total triterpenoid content was also found in winter heather and the greatest
in summer flowering heather. It was also indicated that supercritical CO2 extracts
contain at least 90% of total triterpenoid compared with that of hexane extract. This
means supercritical CO2 at 35 MPa and 50 oC has almost the same extracting ability
as hexane with respect to triterpenoid compounds. However a previous study
indicated that supercritical CO2 is able to extract higher concentrations of triterpenoid
than hexane under conditions of higher pressure and temperature (6000 psi 43MPa)
and 100 oC),2 therefore, future studies can be directed at achieving a better yield of
total triterpenoid content in supercritical CO2 extract by varying supercritical
extraction conditions.
114
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
3.6 Optimal harvest and extraction time
According to the total triterpenoid content quantification, the greatest concentration of
total triterpenoid is present in Calluna vulgaris during August. This would be
naturally the preferred month for harvesting. However, there are several problems
about harvesting heather at this time of the year. The grouse shooting season starts on
12th August every year2 and this is a major source of revenue for North York Moor
National Park and landowners. Harvesting during August will seriously disturb this
sport. August is also the biggest tourist season in the moor, tourism will be greatly
affected if the harvesting takes place in August. Therefore, August would not be the
ideal time to harvest heather when taking these factors into account. Meanwhile, the
burning of heather takes place in spring every year since it is restricted by law that
burning can only take place between 1st October and 15th April.2 In March heather
sample, total triterpenoid content was evaluated as 11200 g/g dry plant in
supercritical CO2 extract and 12400 g/g dry plant in hexane extract. The total
triterpenoid content in March is approximately 88% of that of August heather sample.
Therefore, if harvesting can take place in March instead of burning, the heather will
already contain almost 90% of the highest triterpenoid content in the year. A large
amount of triterpenoid can be extracted if large-scale harvesting of heather can be
achieved at this time of the year.
3.7 Conclusion and further remarks
Seasonal variation of total triterpenoid compounds of Calluna vulgaris supercritical
carbon dioxide extracts was highlighted in this work. The greatest total triterpenoid
content was found in August flowering heather, and the lowest was obtained in
January winter heather. The total triterpenoids in supercritical CO2 with ethanol
entrainer extracts were slight higher than that of supercritical CO2 extracts. Hexane
extracts also showed slightly higher concentrations of total triterpenoids compared to
115
Chapter 3
Seasonal variation of total triterpenoid compounds in heather
supercritical CO2 extracts. However, by varying extraction conditions, it is expected
that supercritical CO2 can extract greater concentrations of triterpenoids than hexane.
Therefore, future work will be done to enhance the extraction of triterpenoids by
using advanced supercritical CO2 extraction conditions.
The optimal harvesting time of heather was also discussed in this chapter. Although
August heather contains the highest level of total triterpenoid, there are difficulties of
harvesting heather in large-scale at this time of the year. Due to grouse shooting and
tourism taking place in August, it is suggested that March would be a good alternative
time to harvest heather since it already contains a high level of total triterpenoid
compounds. Burning of heather is originally taking place in spring every year, so if
harvesting can take place in March instead of burning, the heather will already contain
almost 90% of the highest triterpenoid content in the year. A large amount of
triterpenoid can be extracted if large-scale harvesting of heather can be achieved at
this time of the year.
116
Chapter 4
Extraction and quantification of heather phenolic compounds
Extraction and quantification of
heather total phenolic compounds
Chapter 4
117
Chapter 4
Extraction and quantification of heather phenolic compounds
4.1 Extraction and quantification of heather phenolic compounds
Phenolic compounds are commonly found in both edible and non-edible plants and
are known as one of the largest and most widespread groups of plant secondary
compounds.80 They have been reported to have multiple biological effects, including
antimicrobial activity and antioxidant activity40,43 and phenolic acids have also been
shown to inhibit the in-vitro growth of an assortment of fungal genera.81 Other
workers have shown that single groups of plant phenolics and flavonoids have
applications as antibiotics, anti-ulcer and anti-inflammatory agents.17 Corrales and
co-workers43 evaluated the antimicrobial activity of total phenolic compounds in
grape seed extract. They observed that the growths of Gram-positive food-borne
pathogens were inhibited by phenolic compound, while Gram-negatives were not
inhibited.
Previous work with heather demonstrated a high total phenolic content and
antioxidant activity throughout the year.2 A strong correlation was discovered
between phenolic content and antioxidant activity.2 Heather was found to have the
strongest antioxidant activity in August, a highest total phenolic content was also
detected in this summer flowering sample. The temporary accumulation of phenolic
content in heather during the flowering stage may be due to the use of these
compounds to attract pollinating insects.2
In this chapter, a seasonal variation of total phenolic compound in heather was
examined and phenolic samples at various concentrations were made up for
antimicrobial assay. Qualitative and quantitative test of phenolic antimicrobial activity
were accomplished using several Gram-positive and Gram-negative microorganisms.
118
Chapter 4
Extraction and quantification of heather phenolic compounds
4.2 Seasonal variation of total phenolic compound in heather
Heather whole plant sample picked in October 2007, January 2008, March 2008 and
August 2008 were selected as representative of heather in every season. Heather
flower sample picked in October 2007 was selected as representative of heather
flower. All the raw material was extracted by ethanol follow the method stated in
7.3.5.1. Total phenolic content of each extract was determined using Folin-Ciocalteau
reagent (method described in 7.3.6). Gallic acid was used as a standard for calibration
curve (0-120 g/ml). The total phenolic content was reported as gallic acid
equivalents (GAE)/g dried weight of plant material using the following linear
equation based on the calibration curve;
Y=0.0094X + 0.0056
where Y is the absorbance and X is gallic acid equivalent g/ml. Figure 4.1 shows the
standard curve of gallic acid and the regressive equation. A correlative coefficient of
0.9996 was obtained from this curve.
1.2
absorbance at 765nm
1
y = 0.0094x + 0.0056
R² = 0.9996
0.8
0.6
0.4
0.2
0
0
-0.2
20
40
60
80
100
120
140
gallic acid concentration g/ml
Figure 4.1 The standard curve of gallic acid for heather total phenolics quantification
119
Chapter 4
Extraction and quantification of heather phenolic compounds
Table 4.1 listed the examined yield, absorbance at 765 nm and calculated total
phenolic content of each extract. Figure 5.2 shows the seasonal variation trend of total
phenolic content in these five samples.
Table 4.1 Extract yield and total phenolic content in each heather sample
Sample date
Yield (%)
Total phenolic content in dry plant
October 2007
18.77%
2728 mg/100g
January 2008
21.97%
3750 mg/100g
March 2008
21.42%
5163 mg/100g
August 2008
19.27%
4201 mg/100g
October2007 (flowers)
15.88%
2130 mg/100g
Total phenolic content (mg/100g)
6000
5000
4000
3000
2000
1000
0
10/2007
flower
10/2007
01/2008
03/2008
08/2008
Months
Figure 4.2 Seasonal variation of total phenolic content in heather
The lowest total phenolic content was found in heather flowers, which is 2130
mg/100g dry weight. Heather whole plant seem to have the lowest total phenolics in
120
Chapter 4
Extraction and quantification of heather phenolic compounds
winter and then keep rising until reach a peak in spring sample at 5163 mg/100g dry
plant. Afterwards the total phenolic content drops slightly to 4201 mg/100g dry plant
in summer when it is flowering. The seasonal variation trend is generally very similar
to the previous study, although the greatest total phenolic content in this case was
detected in March sample rather than August.
4.3 Qualitative antimicrobial assay using heather phenolic extracts
Heather sample picked in different seasons (October 2007, January 2008, March 2008
and August 2008) were extracted by ethanol follow the method stated in chapter
7.3.5.1. The crude extracts were then tested for their qualitative antimicrobial activity
on both Gram-positive and Gram-negative bacteria. Paper discs are employed as
reservoirs of heather phenolic extract. Agar plates and bacteria inoculums were
generously provided by department of biology, University of York.
The initial bacteria inoculums were first analysed by a UV spectrophotometer at 500
nm, then the degree of dilution was decided by comparing the absorbance with a
graph of viability for the particular strain. A final concentration of 5 106 cfu/ml was
prepared for each bacterial suspension.
Staphylococcus aureus was selected as representative of Gram-positive cocci, and
Escherichia coli k12 was selected as representative of Gram-negative bacteria. Each
bacterial suspension in right concentration was spread evenly by a cotton swab onto a
sterilized agar plate. Afterwards, sterilized paper disc were dipped into each heather
phenolic extract until the paper disc is saturated. Each paper disc was then placed on
to inoculated agar plate and was pressed until the surface of the disc is fully in contact
with the plate. Since the heather raw materials were extracted by water/ethanol
mixture, deionised water and anhydrous ethanol were also tested as controls. After
121
Chapter 4
Extraction and quantification of heather phenolic compounds
24-hour incubation at 37 oC, inhibition area diameters were recorded and the plates
were photographed.
10/07 flower
10/07
whole plant
Ethanol
control
Deionized
water control
Figure 4.3: Effect of 10/07 flower, 10/07 whole plant,
ethanol control and water control on S. aureus
3/08 heather
8/08 heather
Deionized
water
control
Ethanol
control
1/08 heather
Figure 4.4: Effect of 1/08, 3/08, 8/08, ethanol control and
water control on S. aureus
122
Chapter 4
Extraction and quantification of heather phenolic compounds
1/08 heather
8/08 heather
Deionized
water control
Ethanol
control
10/07 whole
plant
Figure 4.5: Effect of 1/08, 8/08, 10/07whole plant, ethanol
control and water control on E. coli
3/08 heather
Deionized
water
control
10/07 flower
Ethanol
control
Figure 4.6: Effect of 10/07 flower, 3/08, ethanol control and
water control on E. coli
123
Chapter 4
Extraction and quantification of heather phenolic compounds
Table 5.2 listed the antimicrobial activity of each heather ethanolic extract evaluated
by their inhibition zone diameter. From the result, it is clear that all the extracts are
effective to S. aureus and have no inhibition to E. coli. This corroborates the finding
obtained in previous research3 that phenolic compounds inhibit Gram-positive bacteria
but not Gram-negative. Among all the extracts, heather flower exhibited strongest
antibacterial effect with S. aureus and its inhibition zone is 6.0 mm. October whole
plant extract was found to have second greatest bacteria inhibition zone which is 4.6
mm. Winter heather 1/08 sample and spring heather 3/08 sample both showed weak
antibacterial effect and have an inhibition zone of 3 mm. Flowering summer heather
8/08 seemed to have the weakest inhibition to S. aureus and only exhibited a
inhibition zone of 1.5 mm.
Table 4.2:
Antibacterial activity of heather phenolic extracts against common
Gram-positive and Gram-negative bacteria
Inhibition zone (diameter in mm):
0 (-), 1–3 (+), 4–6 (++), 7–10 (+++).
10/07
10/07 whole
1/08
3/08
8/08
flower
plant
S. aureus





E. coli





Water control





Ethanol control





A seasonal variation tendency of antimicrobial activity was showed in figure 4.7. The
trend is a clear reduction from winter to summer. In general, the seasonal variation of
antimicrobial activity is surprisingly in complete contrast with that of the total
phenolic content. Heather flower, which contains the lowest total phenolics, showed
124
Chapter 4
Extraction and quantification of heather phenolic compounds
the strongest antimicrobial activity in this qualitative test. On the other hand, summer
heather sample which have the second highest total phenolic content, exhibited the
weakest antimicrobial activity. The inhibition zone of heather flower was
approximately twice of that of March heather which contains the highest amount of
total phenolic compound.
Inhibition diameter (in mm)
6
5
4
3
2
1
0
10/07 flower
10/2007
01/2008
03/2008
08/2008
Month
Figure 4.7: Seasonal variation of antimicrobial activity of heather
phenolic extracts
When Kahkonen and co-workers2 evaluated the antioxidant activity of 92 plant
ethanolic extracts containing phenolic compound, they found that the antioxidant
activity does not necessarily correlate with high amounts of phenolics, and that is why
both phenolic content and antioxidant activity information must be discussed when
evaluating the antioxidant potential of an extract. Similarly, the antimicrobial activity
does not necessarily correlate with total phenolic content in heather. However, the
reason for why the variation tendency of antimicrobial activity is in contrast to that of
total phenolics content still need more investigation since the composition of total
phenolics in heather was not analysed in this study. Further work is required to
125
Chapter 4
Extraction and quantification of heather phenolic compounds
identify the phenolic compounds present and the antimicrobial activity for each
constituent needs to be evaluated.
4.4 Quantitative antimicrobial test of heather flower phenolic extract
Since heather phenolic extracts are proved to be only effective against Gram-positive
bacteria, a quantitative antimicrobial test was carried out using S. aureus. Heather
flower was selected as extraction material due to its strongest antimicrobial activity as
shown in qualitative assay. Heather flower ethanolic extract was prepared following
the method described in 7.3.5.2. Staphylococcus aureus strain was provided by
Department of Biology, University of York. The initial bacteria inoculum was diluted
by the same method mentioned in qualitative test. A final concentration of 5 106
cfu/ml was prepared for S. aureus bacterial suspension.
Since phenolic compounds are weak acid, there is a possibility that the pH value of
the culture environment may affect its antimicrobial activity. Thus, nutrient broths in
different pH were prepared by adjusting the original broth (pH 7) using citric acid.
Double concentrated nutrient broth was made up by adding 26 g nutrient broth
powder (obtained from Fluka) into 1 L distilled water and mixed well. Then the broth
solution was autoclaved at 120 oC for 15 min prior to use. Nutrient broth at pH 4, 5, 6,
7 and 8 were prepared.
A 96 multiwell culture plate was employed in this study. Figure 4.8 shows the image
of the plate. It contains twelve columns times eight rows of small wells, each one has
a capacity of approximately 200 l.
At the start of the antimicrobial activity assay of heather flower phenolic extract, each
well on the plant was filled up with 100 l double concentrated broth, 2 l bacteria
inoculum and 100 l heather flower phenolic extract. Deionized water and anhydrous
126
Chapter 4
Extraction and quantification of heather phenolic compounds
2% ethanol water solution were also applied as controls. The plate was incubated at
37 oC for 24 hours, and then its turbidity was measured by a microbiophotometer at
580 nm.
Figure 4.8: Round bottomed 96 multi-well culture plate
However, an unexpected situation occurred after incubation. The phenolic compound
in heather flower reacted with the protein content in the broth and produced
precipitate. Thus the reading of each well does not precisely reflect the growth of
bacteria. To solve the problem, 10 ul solution was transferred from each well and
spotted onto sterilized agar plate to give a visible result of the bacteria growth first.
Further action will be taken to quantify the inhibition if any bacteria colonies are
127
Chapter 4
Extraction and quantification of heather phenolic compounds
revealed on this spotted agar plate. The agar plate was incubated at 37 oC for another
24 hours.
pH 4
phenolic
+bacteria
pH 6
phenolic
+bacteria
pH 5
phenolic
+bacteria
pH 7
phenolic
+bacteria
pH 8 phenolic
+bacteria
Figure 4.9: The bacteria colonies of pH 4-pH 8 heather
phenolic extracts
pH 4 water
+bacteria
pH 5 water
+bacteria
pH 6 water
+bacteria
pH 7 water
+bacteria
pH 8 water
+bacteria
Figure 4.10: The bacteria colonies of pH 4-pH 8
water control
128
Chapter 4
Extraction and quantification of heather phenolic compounds
Figure 4.9 and 4.10 show the spotted agar plates of heather phenolic extracts and
water control at different pH values with S. aureus inoculum. Normal colonies are
only revealed in pH 7 in both water control and phenolic solution. This indicate that
the absence of bacteria colonies in other pH value culture is because the bacteria itself
cannot tolerate lower or higher pH environment, rather than the effect of phenolic
antibacterial property. Therefore, another Gram-positive bacteria strain with better pH
tolerance should be selected to replace S. aureus.
To address this situation, lactobacillus was used as experimental subject instead of S.
aureus. Lactobacillus brevis DSM1267 was supplied by Department of Biology,
University of York, and the bacteria suspension was diluted to a final concentration of
5 106 cfu/ml prior to use. Double concentrated MRS broth (obtained from Fluka) was
prepared by adding 102 g broth powder into 1L deionized water and then autoclaved
at 120 oC for 15 min. Moreover, to solve the problem of precipitation, the heather
flower phenolic extract was diluted to contain 1000 ppm and 100 ppm total phenolics
after dilution and prepared follow the method stated in 7.3.5.2. As with the previous
experiment, a 96 multiwell culture plate was used and each well was filled up with
100 l double concentrated MRS broth, 2 l lactobacillus inoculum and 100 l 1000
ppm (or 100 ppm) heather flower phenolic extract. 1% ethanol solution, 0.1%ethanol
solution was also employed as controls. The plate was then incubated at 30 oC for 36
hours. The turbidity of each well was then measured by a microbiophotometer at 485
nm.
Table 4.3 and 4.4 shows the plate arrangement and the final reading of turbidity of
each well. Table 4.5 shows the average turbidity readings of each column of
quadruplicate samples and the inhibit effect of heather flower phenolic extract in 1000
ppm and 100 ppm. The inhibition percentage was calculated based on the bacteria
growth difference in water control and in phenolic extracts.
129
Chapter 4
Extraction and quantification of heather phenolic compounds
Table 4.3: The arrangement of the plate
1
A
B
C
D
2
P1000=1000 ppm phenolic
solution
MRS+P1000
MRS+P1000
+lactobacillus
MRS+P1000
MRS+P1000
+lactobacillus
MRS+P1000
MRS+P1000
+lactobacillus
MRS+P1000
MRS+P1000
+lactobacillus
3
4
5
P100=100 ppm phenolic
solution
MRS+P100
MRS+P100
+lactobacillus
MRS+P100
MRS+P100
+lactobacillus
MRS+P100
MRS+P100
+lactobacillus
MRS+P100
MRS+P100
+lactobacillus
Water control (deionized water instead of phenolic in wells)
E
MRS+H2O
F
MRS+H2O
G
MRS+H2O
H
MRS+H2O
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
MRS+H2O
MRS+H2O
MRS+H2O
MRS+H2O
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
MRS+H2O
+lactobacillus
130
6
EtOH=1% ethanol control
MRS+EtOH
MRS+EtOH
MRS+EtOH
MRS+EtOH
MRS+EtOH
+lactobacillus
MRS+EtOH
+lactobacillus
MRS+EtOH
+lactobacillus
MRS+EtOH
+lactobacillus
7
8
EtOH=0.1% ethanol control
MRS+EtOH
+lactobacillus
MRS+EtOH
MRS+EtOH
+lactobacillus
MRS+EtOH
MRS+EtOH
+lactobacillus
MRS+EtOH
MRS+EtOH
+lactobacillus
MRS+EtOH
Chapter 4
Extraction and quantification of heather phenolic compounds
Table 4.4 Turbidity readings at 485nm of each well after incubation
1
2
P1000=1000 ppm phenolic
solution
3
4
5
P100=100 ppm phenolic
solution
6
EtOH=1% ethanol control
7
8
EtOH=0.1% ethanol control
A
0.485
0.917
0.338
0.881
0.584
0.891
0.465
1.201
B
0.598
0.897
0.476
0.804
0.505
0.858
0.473
1.066
C
0.492
0.912
0.448
0.774
0.342
0.868
0.368
0.929
D
0.366
0.912
0.347
0.786
0.412
0.888
0.627
0.927
Water control (deionized water instead of phenolic in wells)
E
0.422
0.836
0.358
0.800
F
0.444
0.814
0.472
0.832
G
0.380
0.911
0.342
0.784
H
0.361
0.928
0.477
1.032
131
Chapter 4
Extraction and quantification of heather phenolic compounds
Table 4.5 Average turbidity readings of each column and calculated inhibition effect of phenolic extracts
Average turbidity at 485nm
1000ppm phenolic
A: broth + phenolic
extract without bacteria
after incubation
(blank readings)
B: broth+phenolic
extract with bacteria
inoculums after
incubation
Total growth: B-A
0.48525
0.90950
0.42425
Broth with water
control
0.40175
0.87225
0.47050
100ppm phenolic
0.40225
0.81125
0.40900
Broth with water
control
0.41225
0.86200
0.44975
1%EtOH control
0.44500
0.89013
0.44513
0.1%EtOH control
0.47163
0.98663
0.51400
Inhibition
P1000=1000 ppm
phenolic solution
P100=100 ppm
phenolic solution
P1000 effect
-0.04625
9.83%
P100 effect
-0.04075
9.06%
0%
132
Chapter 4
Extraction and quantification of heather phenolic compounds
After 36 hours incubation, 1000 ppm heather phenolic extract and 100 ppm phenolic
extract inhibited 9.83% and 9.06% of total growth of lactobacillus respectively. These
inhibition figures are very low compared to other commonly used antimicrobial
agents. Therefore, this indicated that heather flower phenolic extract at 1000 ppm and
100 ppm both have very weak antimicrobial activity and are not sufficient to be
applied as antimicrobial agent even at these fairly high concentrations. However,
since the composition of heather flower phenolic were not analysed in this study, it is
worth being investigated further to identify the constituent of heather flower total
phenolics. Thus, any key components which express noticeable antibacterial activity
can be purified and evaluated for their quantified antimicrobial effect to raise the
possibility of heather being used as an herbal antibacterial.
4.5 Conclusion and further remarks
Heather plant picked in October 2007, January 2008, March 2008 and August 2008
were selected as representative of heather in every season. Heather flower picked in
October 2007 was selected as representative of heather flower. Each heather raw
material was extracted by ethanol/water solution and was then evaluated for their total
phenolic content by using UV spectrophotometer. Gallic acid was employed as a
standard in this quantification and the total phenolic content was reported as gallic
acid equivalents (GAE)/g dried weight of plant material. A seasonal variation trend
was obtained and a clear increasing of total phenolic content in heather from autumn
to spring was showed. Heather flower is proved to have the lowest total phenolic
content which is 2130 mg/100g dry plant, and spring heather contains the highest total
phenolic content at 5163 mg/100g dry plant.
However, when the same heather ethanolic extracts were evaluated for their
qualitative antimicrobial activity, an obvious contrast was observed between phenolic
content and inhibition of gram positive bacteria. Paper disc qualitative antimicrobial
133
Chapter 4
Extraction and quantification of heather phenolic compounds
test was used and the inhibiting effect of each heather extract was estimated by
inhibition zone diameter. Heather phenolic extract was proved to be only effective on
Gram-positive bacteria which confirmed the results obtained in previous studies.
Heather flower with the lowest total phenolic content showed the largest inhibition
zone, and summer heather which contains the second highest total phenolic content
showed the weakest antimicrobial ability. This phenomenon proves that the
antimicrobial activity does not necessarily correlate with high content of total
phenolics. Therefore, further investigation is required to identify the composition of
heather total phenolic to give a better explanation.
Heather flower phenolic extract was then analysed for its quantitative antimicrobial
activity. Multiwell culture plate was employed and S. aureus was selected as
experimental subject first. Since phenolic content is weak acid, different pH broths
were prepared to estimated the effect of pH value on phenolic antimicrobial activity.
However, after 24 hours incubation, the phenolic content was found to have reacted
with the protein in the broth and generated precipitate that made the turbidity reading
less precise. To solve the problem, 10 l solution from each well was spotted to
sterilized agar plate to give visible bacteria colonies before further quantification.
After another 24 hours incubation at 37 oC, S. aureus was found only survive in pH 7
broth in both water control and phenolic solution. A more tolerant Gram-positive
organism was selected to give better growth in lower pH conditions.
Lactobacillus brevis was then applied instead of S. aureus, and the precipitation
problem was resolved by diluting heather phenolic content to 1000 ppm and 100 ppm.
96 multiwell culture plates were again used and the plate was incubated at 30 oC for
36 hours after inoculation. The turbidity of each well was then measured by a
microbiophotometer at 485 nm. After calculation, heather phenolic extract inhibited
9.83% of total bacteria growth at 1000 ppm and 9.06% at 100 ppm. These figures are
very low compared with commonly used antimicrobial agent. To conclude, heather
flower phenolic extract did not show strong antimicrobial activity at 1000 ppm and
134
Chapter 4
Extraction and quantification of heather phenolic compounds
100 ppm. However, since the composition of heather flower phenolics was not
analysed in this study, it is worth being investigated further to identify the constituent
of heather flower total phenolics. Key components with better antibacterial activity
can be purified and evaluated for their quantified antimicrobial effect to raise the
possibility of heather being used as herbal antibacterial.
135
Chapter 4
Extraction and quantification of heather phenolic compounds
136
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Liquid and supercritical carbon
dioxide extractions of bracken
Chapter 5
137
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5.1 Liquid CO2 and supercritical CO2 extractions of bracken
The application of supercritical fluid extraction especially liquid and supercritical
carbon dioxide is getting more and more attention in the field of natural product
research. This environmentally benign extraction technique offers extraction yields
that are comparable with those achieved by conventional solvent extraction.
Meantime, the remarkable selectivity of liquid and supercritical CO2 has unique
superiority and makes this technique a perfect separating method. Moreover, in
contrast with organic solvents, carbon dioxide is non-toxic, non-flammable,
non-corrosive, cheap and readily available in large quantities with high purity.82 Since
the critical pressure (72.8 bar) and critical temperature (31 oC) of carbon dioxide is
easily accessible, liquid and supercritical CO2 are really ideal solvent for natural
compound extraction.83
To achieve a comprehensive scan of the compounds present in bracken, liquid CO2
extraction was employed along with supercritical CO2 extraction. Moreover, to
enhance to solvent power of the supercritical CO2, the use of ethanol entrainer was
also introduced in this work. Thus, a standard bracken extraction scheme is first,
liquid CO2 extraction, followed by supercritical CO2 extraction and finally
supercritical CO2 with ethanol entrainer. To the best of my knowledge, liquid and
supercritical CO2 extractions have never carried on bracken plant before. Bracken was
normally either extracted by organic solvent or water due to the solubility of
ptaquiloside in aqueous solution. Therefore, this work is of great meaning to develop
and probe a new field of knowledge of the plant. By analysing each extract obtained
from these consecutive processes, a panorama of bracken chemical compounds could
be revealed.
Samples in different bracken life cycle steps are picked from the same site (Kilburn
White Horse, Sutton Bank, Grid reference SE 514 813) in different months of a year.
138
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
The first crozier sample was picked in 17th April 2009 and 19th May 2010, followed
by the 2nd step sample picked in 21st May 2009, 3rd sample picked in 27th July 2009
and 4th sample picked in 12th October 2009. Unfortunately, the sample picked in
April 2009 was misidentified and found to be lady fern rather than bracken fern, all
the data and information for bracken crozier was obtained from May 2010 sample. All
the plant samples were dried in room temperature and milled to a particle size of ≤2
mm.
5.2 Liquid and supercritical carbon dioxide extraction of bracken
crozier
5.2.1 Liquid CO2 extraction of bracken crozier
It was reported that the carcinogenicity of young bracken shoots was shown to be
much stronger than that of mature plant.6 Due to the remarkable distinction of bracken
crozier, liquid CO2 extraction, along with supercritical CO2 extraction was employed
in bracken crozier analysis to give a more comprehensive understanding of its
chemical content. The conditions for bracken liquid carbon dioxide extraction were
set to 5 oC and 65 bar and a CO2 flow-rate of 40 g/min. An initial yield of 0.60% was
obtained from this extraction. The whole liquid CO2 extraction lasted for 6 hours, and
the extract was recovered from the separator by washing out with dichloromethane
then concentrated by removing the solvent using a rotary evaporator. It seems liquid
CO2 is able to extract more moisture from the raw material than supercritical CO2
since the extract was moister than the waxy extract obtained by supercritical
extraction. A sample with a concentration of 20 mg/ml of crude extract was made up
for GC-MS analysis.
139
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5.2.1.1 Identification of pterosin B
Figure 5.1 shows the gas chromatogram of bracken liquid CO2 extract and also
highlights the composition. It is well known that ptaquiloside presents in bracken
young frond by a very high amount, but the quantification of ptaquiloside is really
difficult due to its nature of instability. Therefore pterosin B, known as the most
common final metabolite of ptaquiloside, is usually used as a representative
compound in ptaquiloside quantification. Pterosin B (C14H18O2, MW 218) has a basic
indanone structure, and can be generated directly from ptaquiloside decomposition. In
bracken liquid CO2 extract, pterosin B appeared at 21.02 min and its identification
was achieved by GC-MS analysis.
140
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
lz bracken
Bracken5/10
5/10liquid
liquidco2
COextract
extract
, 10-Mar-2011 + 14:03:43
2
lz-blco2510k
Scan EI+
TIC
5.90e8
33.42
100
-sitosterol
%
Pterosin B
Pterosin F
20.32 21.02
Long chain primary alcohols
38.85
heptacosane
3.02
32.71
n-hexadecanoic acid
33.55 34.92
40.26 41.93
44.34
46.95
37.63
27.69
22.34
0
Time
5.50
10.50
15.50
20.50
25.50
30.50
Figure 5.1 Gas chromatogram of bracken crozier liquid CO2 extract
141
35.50
40.50
45.50
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Figure 5.2 shows the mass spectrum of pterosin B. Major ions are observed at m/z
187, 203 and 218. Benzylic cleavage of the side chain is an important process that
gives rise to the ion at m/z 187, and the abundance of m/z 203 is due to the loss of
methyl group. These characteristic ions lend support to the previous spectral data
obtained by Bonadies and co-workers.84
bracken 5/10 kilburn SCCO2 extract
, 22-May-2010 + 10:09:30
[M-CH2OH]+
Mass spectrum of pterosin B
lzbsc510k 2606 (20.396) Cm (2602:2609-2621:2626)
Scan EI+
2.61e7
187
100
CH3
O
[M-15]+
HO
CH3
H3C
%
203
[M-CH2CH2OH]+
M●+
175
128
115
218
91
77
65
51
53 55
63
76 79
44 50
58
41
69 70
43
0
105
117
127
119
103
87 89 92 95 101
106 114
82
144
141
145
131
188
159
157
132
126
134 139
147
173
160
152 154
171
167
204
176
177
185
183
189
190
199 201
219
205
214 216
m/z
41
51
61
71
81
91
101
111
121
131
141
151
161
171
181
191
201
211
221
Figure 5.2 Mass spectrum of pterosin B
Summary of GC/MS data for compounds III-VIII
GC retention times and MS ions
Pterosin B (III)
GC (min)
GC/MS (min)
m/z (abundance)
7.89
12.66
218(42); 203(85); 187(100)
Figure 5.3 GC-MS data of pterosin B published by Bonadies84
142
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
+
.
•
-CH3
m/z 203
•
-CH2CH2OH
m/z 218
m/z 175
•
-CH2OH
m/z 187
-
-CO
-
m/z 159
Scheme 5.1: Proposed mechanism for
fragmentation of pterosin B
143
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Since the conditions of liquid CO2 extraction were quite mild, the pterosin B present
in this extract is more likely to be in the plant originally prior to the extraction. As
mentioned before, pterosin B is usually used as a representative of ptaquiloside. But
there is a consideration that the transformation of ptaquiloside to pterosin B would
already occur in the frond when alive since ptaquiloside is known to degrade into
pterosin B under heat, acid or base. Therefore, the identification of ptaquiloside using
pterosin B may not really reflect the presence of ptaquiloside in the plant matrix. This
discovery of pterosin B in liquid CO2 extract confirms this supposition. The presence
of pterosin B in this extract means that the decomposition of ptaquiloside already
exists during the early growing stage of the frond, and the quantity of pterosin B after
experimental conversion is a summation of this compound present originally in the
plant and pterosin B converted from ptaquiloside during the experiment.
5.2.1.2 Identification of pterosin F
Along with pterosin B, pterosin F (C14H17OCl MW 236.5) is also detected in this
extract. Pterosin F has a very similar structure with pterosin B, only with chlorine
instead of the hydroxyl on the side chain. Pterosin F peak exist at 20.57 min, same as
pterosin B, its identification is also achieved by GC-MS analysis. Figure 5.4 shows
the mass spectrum of pterosin F.
144
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5/09 bracken
SCCO2
extract
Mass
spectrum
of pterosin F
, 26-Apr-2010 + 13:01:53
lzbscco2-509 2563 (20.129) Cm (2561:2565-2574:2579)
[M-Cl]+
100
Scan EI+
1.18e7
201
[M-CH2Cl]+
187
M●+
%
128
236
91
43
53 55
41 44 49
141
77 80
51
63 65
58
116
76
85
83 89 93
69 71
[M-15]+
129
115
99
105
107 114
127
159
144
131
119 126
132 139
146
155 157
160
50
60
70
80
90
100
110
120
130
140
150
181
160
170
238
221
185
168 172 174
0
40
202
188
180
189 193 199 203
190
200
Figure 5.4 Mass spectrum of pterosin F
Figure 5.5 Mass spectra of pterosin F published by
Bonadies84
145
213
210
235
223
239
m/z
220
230
240
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
+
.
•
-CH3
m/z 221
•
-Cl
m/z 236
m/z 201
•
-CH2Cl
m/z 187
Same process
as pterosin B
-
m/z 159
Scheme 5.2: Proposed mechanism for
fragmentation of pterosin F
Characteristic ions were found at m/z 187, 201 and 236. As seen in pterosin B,
benzylic cleavage of the side chain is the main reason for the ion at m/z 187. The loss
of the chlorine from the side chain affords the ion at m/z 201. According to Bonadies
and co-workers, this ion only dominates when the substituent on the side chain can be
146
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
released as low-energy radical. In the case of pterosin B, this mechanism is not
favorable, thus the loss of methyl group become more significant and affords an ion at
m/z 203.84 The molecular ion falls at m/z 236, and the ratio between m/z 236 and m/z
238 precisely reflects the relative abundance of chlorine isotopes (35Cl:37Cl 3:1).
Scheme 5.2 shows the proposed mechanism for fragmentation of pterosin F.
The presence of pterosin F in this extract also indicates that it already exists in the
frond prior to the extraction. In previous studies, pterosin F was found in methanol
extract of young leaves.85,86 However, it was never reported whether pterosin F is
present in the frond alone or as a metabolic product from a related precursor. As
suggested by Saito, pterosin B results from light-induced decomposition of
ptaquiloside.87 Therefore, it is conceivable that ptaquiloside is a relocalizable source
of pterosin B in the frond. Meanwhile, another indanone compound, pterosin A, exists
in bracken and was suggested to have its own ptaquiloside-like precursor which was
isolated from another fern species; although this precursor named dennstoside A has
not been discovered in bracken fern.88 In the case of pterosin F, any precursor or
transformation mechanism is currently unknown. Further study is required to
determine the source of pterosin F in bracken.
5.2.1.3 Identification of other major compounds found in bracken crozier liquid
CO2 extract
Some other compounds are also found in this liquid CO2 extract. -sitosterol(C29H50O
MW 414) is a widely distributed phytosterol that is found in many plant species. It’s
well known for its blood cholesterol lowering effect. -sitosterol limits the amount of
cholesterol entering the body by inhibiting cholesterol absorption in the intestines,
therefore decreasing the levels of cholesterol in the body. It is also helpful with benign
prostatic hyperplasia (BPH), due to its anti-inflammatory effects and its ability to
improve urinary symptoms and flow.89 The -sitosterol peak appears at 33.42 min,
147
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
and its quantification was also achieved by using stigmasterol as a standard. Figure
5.6 shows the standard spectra of -sitosterol and spectra of peak at 33.42 min.
+
.
•
-C10H21 side chain
A
m/z 273
-H2O
B
m/z 255
+
.
•
-C13H26-H
C
m/z 231
-H2O
D
m/z 213
+
.
•
-CH3
m/z 399
A ring
cleavage
-C4H8O+2H
E
m/z 329
148
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
+
•
+
•
-H2O
F
m/z 396
•
-CH3
G
m/z 381
Scheme 5.2: Proposed mechanism for
fragmentation of -sitosterol
149
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
C
D
100
A
B
E
F
G
43
50
HO
55
81
107
414
95
69
133
145
161
119
213
85
173
255
231
187 199
241
0
40
60
(mainlib) β-Sitosterol
80
100
120
140
160
180
200
220
240
303
381
283
260
396
329
273
280
314
300
320
341
340
354
360
387
380
400
Figure 5.6 Mass spectra of -sitosterol and the peak with retention time at 33.42min
150
420
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Stigmasterol(C29H48O, MW 412) is another phytosterol with a structure very similar
to that of -sitosterol. Thus it can be used as a standard for -sitosterol quantification
since their response factors would be very similar. Figure 5.7 shows the chemical
structure of stigmasterol and -sitosterol.
Standard solutions of stigmasterol in 0.6, 1.0, 1.5, 1.9, 2.5 and 3mg/ml concentrations
were made up and analysed by GC. The absolute areas were recorded and a standard
curve was determined using these area data. Figure 5.8 shows the standard curve of
stigmasterol and the regressive equation. A correlative coefficient of 0.994 was
obtained from this curve.
-sitosterol
stigmasterol
Figure 5.7 Chemical structures of stigmasterol and -sitosterol
151
Millions
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
18
y = 6E+06x - 1E+06
R² = 0.9946
16
Area of peak (pA*s)
14
12
10
8
6
4
2
0
0
0.5
1
1.5
2
2.5
3
3.5
Concentration of stigmasterol (mg/ml)
Figure 5.8 The standard curve of stigmasterol
for -sitosterol quantification
The -sitosterol content was then calculated from the equation by using its absolute
peak area. The quantity of it was found to be 1217 g/g dry plant and 324 g/g fresh
plant. Its seasonal variation was also examined in later chapter.
Some long chain primary alcohols were also found in this extract. The two peaks with
retention time of 37.63 min and 38.85 min both showed very similar spectra to a long
chain primary alcohol standard (1-triacontanol C30H62O). Figure 5.9 shows the spectra
comparison between the two peaks and 1-triacontanol standard.
152
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
lz bracken 5/10 liquid co2 extract
, 10-Mar-2011 + 14:03:43
lz-blco2510k 5188 (37.627) Cm (5185:5191-5205:5225)
43
100
Scan EI+
1.38e6
57
Mass spectra of the peak at 37.63min
55
%
71 83
97
82
85
68
98 111
79
112 129 139 153
41
44
0
35
55
75
95
115
135
171 185
155
175
201 210 229 240
195
215
235
258
255
296 314
271 286
325
275
295
315
343 356 370
335
355
388 402 416 430 442
375
lz bracken 5/10 liquid co2 extract
395
415
, 10-Mar-2011 + 14:03:43
lz-blco2510k 5371 (38.849) Cm (5368:5374-5385:5435)
43
100
m/z
435
Scan EI+
1.92e6
55
Mass spectra of the peak at 38.85min
%
83
69
41
54
82
85
97
98
111
93
53
125
0
35
55
75
95
115
136 149 161
135
155
179
175
208
191
240 253 263
219
195
215
235
255
275
286 297
295
314 327 341
315
335
359
355
389 402
375
395
m/z
415
435
57
100
OH
43
83
97
69
50
Mass spectra of 1-triacontanol standard
111
125
0
20
40
60
(mainlib) 1-Triacontanol
420
139
29
80
100
120
140
153 167
181 195 209
223 237 251 265 279 292 306 321 336 350 362 376
160
180
200
220
240
260
280
300
320
340
360
380
392
400
437
420
Figure 5.9 Mass spectra of the two peaks in liquid CO2 extract and 1-triacontanol standard
153
440
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
The three spectra are all very similar in respect of the key fragment ions, indicating
that these are a homologous series of long chain primary alcohols. No molecular ion
is present to indicate the chain length; however, the Kovats index of each peak shows
that they are separated by two CH2 units (200 Kovats units). Table 5.1 shows the
published Kovat index of 1- heptatriacontanol and 1-nonatriacontanol with KI value
calculated for the two peaks. The chain length is 37 carbons for the peak at 37.63 min
based on the calculated KI value relative to estimated standard KI data, and for the
peak at 38.85 min the chain length was determined to be 39 carbons. Estimated
standard KI value was calculated based on the standard KI data of heptacosanol and
triacontanol published by Nist Mass Spectrometry Data Centre.
Table 5.1 Kovats index calculated for the two peaks
and comparison with published data
Retention
time
Possible compound
Molecular
formula and
weight
KI value
calculated
Estimated
standard KI
value
37.6min
1-heptatriacontanol
C37H76O
MW 536
3977
3942
38.85min
1-nonatriacontanol
C39H80O
MW 564
4165
4142
5.2.2 Supercritical carbon dioxide extractions of bracken crozier
Bracken crozier supercritical CO2 extraction was carried out at 50 oC, 350 bar and a
flow-rate of 40 g CO2/minute for 4 hours. The extract was recovered by washing out
the separator with dichloromethane and then concentrated by evaporating the solvent
using a rotary evaporator. A light green waxy extract was obtained with a crude yield
of 0.68%. A sample was made up in dichloromethane at a concentration of 20 mg/ml
of crude extract for GC-MS analysis.
154
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Figure 5.10 shows the gas chromatogram of bracken crozier supercritical CO2 extract.
Pterosin B and pterosin F appear once again in this extract at 20.40 min and 19.94 min.
The mass spectra of pterosin B and F are identical to those of the corresponding peaks
in liquid CO2 extract and published spectra.
-sitosterol is also present in this extract and is identified as the peak with retention
time of 32.64 min. The quantification of -sitosterol in this extract was also achieved
by using stigmasterol as a standard. The quantity of -sitosterol was calculated as 754
g/g dry plant and 201 g/g fresh plant.
The chromatogram indicates clearly that more long chain primary alcohols were
extracted using supercritical CO2. In this extract, alcohols with more carbon numbers
were extracted along with the two peaks that have been identified in 5.2.1.3. The
peaks at 36.87 min and 38.07 min in this extract have identical spectra and Kovats
index to the heptatriacontanol and nonatriacontanol that were identified in the liquid
CO2 extract. (The retention time of these two peaks has shifted ahead since the
column of GC-MS has been removed and reinstalled between analyses)
Figure 5.11 shows the spectra of the series of alcohol peaks and 1-triacontanol
standard. As can see from the figure, the series of peaks all have nearly identical
spectra with 1-triacontanol standard with molecular ion missing. Thus, Kovats index
was applied again to help identify the chain length.
155
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
bracken 5/10 kilburn SCCO2 extract
Bracken 5/10 scCO2 extract
lzbsc510k
, 22-May-2010 + 10:09:30
Scan EI+
TIC
3.50e8
20.40
100
32.64
Pterosin B
-sitosterol
Pterosin F
Long chain primary alcohols
%
40.70
19.94
42.47
39.27
44.74
47.63
Phytol and related
19.76
n-hexadecanoic acid
44.42
38.07
derivative
41.82
21.56
21.77
46.05
32.77
31.94
35.48 36.87
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 5.10 Gas chromatogram of bracken crozier supercritical CO2 extract
156
40.51
45.51
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
bracken 5/10 kilburn SCCO2 extract
, 22-May-2010 + 10:09:30
lzbsc510k 5434 (39.274) Cm (5432:5435-5445:5453)
Scan EI+
6.15e6
57
100
43 55
83
69
97
Mass spectra of the peak at 39.27min
85
%
41 54
68
98 111
82
95
125
49
0
35
55
75
95
115
139 153 167
135
155
175
185
191
195
210
215
267
241 258
235
255
275
286
296
295
314
315
331 341 355 370 382
335
355
bracken 5/10 kilburn SCCO2 extract
387
375
395
415 429
415
, 22-May-2010 + 10:09:30
lzbsc510k 5647 (40.697) Cm (5641:5652-5663:5683)
Scan EI+
5.52e6
57
100
m/z
435
43
55
%
42 44
71 83 97
85
82
111
68
96 98
125
73
0
35
55
75
95
115
Mass spectra of the peak at 40.70min
139 151 167
135
155
175
185
191
195
209
215
267
235
255
275
286 296 314
295
315
327 341 355 370 382
335
355
bracken 5/10 kilburn SCCO2 extract
375
417 431
395
415
, 22-May-2010 + 10:09:30
lzbsc510k 6252 (44.739) Cm (6244:6261-6265:6413)
Scan EI+
2.22e6
57
100
m/z
435
43
55
71 83
%
82
42 54
68
Mass spectra of the peak at 42.47min
97
96 98 111
80
125 133
0
35
55
75
95
115
135
207
295 314
227 241 257 269 285
153 167 181 192
155
175
195
215
235
157
255
275
295
315
331 341 356 370
335
355
375
431
395
415
435
m/z
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
bracken 5/10 kilburn SCCO2 extract
, 22-May-2010 + 10:09:30
lzbsc510k 6252 (44.739) Cm (6244:6261-6265:6413)
Scan EI+
2.22e6
57
100
43
55
Mass spectra of the peak at 44.74min
71 83 97
%
82
42 54
68
96 98 111
80
125 133
0
35
55
75
95
115
135
207
295 314
227 241 257 269 285
153 167 181 192
155
175
195
215
235
255
275
295
331 341 356 370
315
335
355
431
375
395
415
m/z
435
57
100
OH
43
83
97
69
50
111
125
0
20
40
60
(mainlib) 1-Triacontanol
420
139
29
80
100
120
140
153 167
181 195 209
223 237 251 265 279 292 306 321 336 350 362 376 392
160
180
200
220
240
260
280
300
320
340
360
380
400
437
420
440
Figure 5.11Mass spectra of long chain alcohol peaks in supercritical CO2 extract and 1-triacontanol standard
158
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Table 5.2 shows the published or estimated Kovats index of standard alcohols and KI
value calculated for the peak series. Estimated standard KI value was calculated based
on the standard KI data of heptacosanol and triacontanol published by Nist Mass
Spectrometry Data Centre. The Kovats index of each peak shows that they are
separated by two CH2 unit (200 Kovats units), and all the peaks have a difference
within 32 Kovats unit with their corresponding standard. The variance between
calculated KI and estimated standard KI may due to the difference of temperature
conditions applied in gas chromatography programs and some of this variance can be
explained by the transition from temperature programmed conditions to isothermal at
this part of the chromatogram. The peaks are identified to be 1-hentetracontanol,
1-tritetracontanol, 1-pentatetracontanol and 1-heptatetracontanol. The presence of
longer chain alcohols in the supercritical extract indicates that the extracting ability of
scCO2 is substantially higher than liquid CO2. With higher temperature and pressure,
supercritical CO2 is able to extract molecules with greater polarity and higher
molecular weight. This result indicates that the selectivity of CO2 is dependent on
operating conditions.
Table 5.2 Calculated KI value of alcohol peak series
and corresponding standard KI value
Retention
time
Possible compound
Molecular
formula and
weight
KI value
calculated
Estimated
standard KI
value
39.27min
1-hentetracontanol
C41H84O
MW 592
4374
4342
40.70min
1-tritetracontanol
C43H88O
MW 620
4535
4542
42.47min
1-pentatetracontanol
C45H92O
MW 648
4727
4742
44.74min
1-heptatetracontanol
C47H96O
MW 676
4970
4942
159
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5.2.3 Supercritical carbon dioxide extraction of bracken crozier with ethanol
entrainer
Bracken crozier was extracted using supercritical CO2 and ethanol entrainer under
conditions of 50 oC, 350 bar for 4 hours and a CO2 flow-rate of 40 g/min. Entrainer
flow rate was controlled at 10% of CO2 flow rate i.e. 4 ml/min. The extract was
washed out of the separator using dichloromethane and then concentrated by
evaporating the solvent using a rotary evaporator. A dark green waxy extract was
obtained with a crude yield of 0.84%. A sample was made up in ethanol at a
concentration of 20 mg/ml of crude extract for GC-MS analysis.
Figure 5.12 shows the gas chromatogram of bracken crozier extracted by supercritical
CO2 with ethanol entrainer. Basically, the pattern is very similar to that of liquid CO2
extract and supercritical CO2 extract. Major peaks like pterosin B and -sitosterol still
exist in the extract. However, another important compound pterosin F does not appear
in this extract anymore and this may due to it having been extracted completely in
previous extractions. The quantity of -sitosterol in this extract was calculated as 702
g/g dry plant and 187 g/g fresh plant. Several new peaks appear in this extract due
to the greater polarity of scCO2 with ethanol entrainer. Some additional branched
unsaturated alcohols related to phytol were detected in this extract, and further
sitosterol derivatives also eluted immediately prior to the -sitosterol peak.
160
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
bracken 5/10 kilburn SCCO2+EtOH extract
, 22-May-2010 + 11:13:16
lzbsc+etoh510k
Scan EI+
TIC
2.53e8
32.64
100
-sitosterol
%
n-hexadecanoic
19.75
acid
Pterosin B
Phytol related
Sitosterol
acetate
20.39
18.27
6.78
30.74
21.57
21.84
16.62
4.72
45.88 47.32 47.99
30.90
branched alcohols
3.47
43.42 44.16
25.62
25.31
9.38
31.93
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 5.12 Gas chromatogram of bracken crozier supercritical CO2 with ethanol extract
161
40.51
45.51
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
When comparing the gas chromatogram of all three bracken crozier extracts, it seems
clear that pterosin B has higher abundance compared to pterosin F. Pterosin B
afforded greater peak areas than pterosin F in every chromatogram. Moreover,
pterosin B presents consistently throughout the extracts, meanwhile pterosin F does
not appear in the last extract anymore. Due to the scarcity of pterosin standards,
quantification of pterosin B and pterosin F was not carried out, thus the comparison of
their quantity in bracken crozier cannot be achieved precisely. However, a seasonal
variation of pterosins in different life stage samples was demonstrated by relative
level comparison with an internal standard and this is described in a later chapter.
5.3 Supercritical carbon dioxide extraction of bracken at different
growth stages
5.3.1 Supercritical CO2 extractions of bracken samples at different growth stages
Bracken samples harvested in May 09, July 09 and October 09 were air dried and
extracted by supercritical carbon dioxide under a condition of 35 MPa and 50 oC for
four hours. After the extraction, the extract was recovered from the separator by
washing out with dichloromethane then concentrated by removing the solvent using a
rotary evaporator. Light green waxy extracts were obtained with crude yields varies
from 1.31% to 1.84% dry plant and 0.39% to 1.02% in fresh plant. Samples of each
extract were made up at a concentration of 20 mg/ml for GC-MS analysis.
Table 5.3 lists the extracted material and yield of each extraction. The figures clearly
stated that the total yield in both dry plant and fresh plant varies in a fairly wide range
with the seasons. The seasonal variation of bracken extraction yield will be discussed
in more detail in a later chapter.
162
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Table 5.3 The extract yields of supercritical carbon dioxide extractions.
Material
Harvest
date
Extraction
conditions
Weight
of extract
Yield
(dry plant)
Yield
(fresh plant)
Dry bracken
whole plant 90g
May
2009
35 MPa, 50 oC,
4 hrs
1.18 g
1.31%
0.39%
Dry bracken
whole plant 70g
July
2009
35 MPa, 50 oC,
4 hrs
1.14 g
1.63%
0.76%
Dry bracken
October
35 MPa, 50 oC,
1.29 g
1.84%
1.02%
whole plant 70g
2009
4 hrs
Figure 5.13 shows the gas chromatogram of bracken supercritical CO 2 extract in
different life stages. The pattern of the May 09 sample is almost the same with that of
bracken crozier. Most interesting peaks such as pterosin B, pterosin F and -sitosterol
all still present in this extract. As was observed in the crozier supercritical CO2 extract,
pterosin B is present at a fairly high concentration and afforded a peak of the greatest
absolute area in this extract. Pterosin F also exhibited a relative high content in May
09 bracken, its peak area achieved approximately 30% of that of pterosin B.
-sitosterol is also one of the major peaks in this extract, its quantity was calculated as
1074 g/g dry plant and 318 g/g fresh plant by using stigmasterol as standard.
The chromatogram of July bracken sample is quite different from earlier ones. As the
season changes, the chemical composition in the plant varies significantly. The most
interesting change is in this fully matured frond, pterosin B and pterosin F exist at
very low concentration compared to younger frond. The peak area of pterosin B in
this mature bracken is only about 3% of that in May sample when the GC samples are
made in the same concentration. Meanwhile, pterosin F peak is not even detectable in
this extract. Moreover, when looking at the chromatogram of October sample, both
pterosin B and pterosin F have disappeared. No peak within the time range of 19.87
min to 21.68 min is detected in this senescing frond. Apparently, there is a clear
seasonal dependent reduction of pterosin compounds in bracken plant. The seasonal
variation of pterosins in bracken will be discussed in detail in later chapter.
163
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5/09
bracken
Bracken
5/09SCCO2
scCO2 extract
extract
, 26-Apr-2010 + 13:01:53
lzbscco2-509
20.57
100
Pterosin B
-sitosterol
Pterosin F
%
Scan EI+
TIC
3.94e8
Long chain primary alcohols
32.86
19.93
39.53 41.01
38.28 40.75
phytol
n-hexadecanoic acid
21.73
32.16
27.26
0
lzbscco2-709
32.99 34.37
100
Phytol and related
pterosins
21.73
40.99
branched alcohols
18.46
13.63
32.41
20.57
27.26
34.39
Scan EI+
TIC
1.57e8
45.26
45.20
39.50
38.28
3.03
42.89
32.86
%
47.75
37.07
-sitosterol
Bracken 7/09 scCO2 extract
42.88 44.58 45.29 45.97
47.01 47.91
42.59
39.42
37.64
29.37
0
lzbscco2-1009
Scan EI+
Bracken 10/09 scCO2 extract
42.76
100
-sitosterol
39.45
32.79
40.93
42.19
TIC
45.10
47.77 3.02e8
44.81 46.02
39.23
%
38.05
18.40 19.87
31.32
21.68
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 5.13: Gas chromatogram of bracken supercritical CO2 extracts
164
40.51
45.51
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
In the chromatogram of July sample, phytol and other two related branched
unsaturated alcohols with retention time of 18.48 min, 21.73 min and 34.39 min
become noticeable. Particularly, phytol peak afforded the second greatest absolute
area in this chromatogram. Meanwhile, -sitosterol still presented as a major peak. Its
quantity was calculated as 660 g/g in dry plant and 309 g/g in fresh plant. Apart
from this, C37-C47 long chain primary alcohols are also more prominent in this
extract. Their spectra and Kovats index values are all identical with standard and
previous calculation.
As mentioned before, pterosin B and pterosin F no longer appear in October bracken
extract. However, other major peaks such as -sitosterol and long chain alcohols still
exist in this extract. -sitosterol was found to have a content of 1132 g/g dry plant
and 625 g/g fresh plant in this over-matured frond.
To conclude, -sitosterol and C37-C47 long chain primary alcohols present
persistently in bracken supercritical CO2 extract at every growth stage. The seasonal
variation trend of -sitosterol in both dry plant and fresh plant will be discussed in
detail later. The amount of Pterosin B and pterosin F in bracken plant seems to
decrease considerably as the season progresses. The supercritical CO2 extract of
younger frond (May10 and May 09 samples) contain relatively higher level of
pterosin than mature frond (July 09 sample) and senescing fronds no longer contain
any pterosin compounds.
5.3.2 Supercritical CO2 extractions of bracken with ethanol entrainer at different
growth stages
Bracken samples harvested in May 09, July 09 and October 09 were air dried as
before and extracted using supercritical carbon dioxide under a condition of 35 MPa
and 50 oC for four hours. The residue from this extraction was then extracted using
supercritical carbon dioxide at 35 MPa and 50 oC with the addition of ethanol as an
165
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
entrainer. The co-solvent flow rate was set to 4 ml/min to make it 10% of the flow
rate of CO2. After the extractions, the extract were recovered from the separator by
washing out with dichloromethane then concentrated by removing the solvent using a
rotary evaporator. Dark green waxy extract were obtained with crude yields varies
from 0.74% to 0.90% dry plant and 0.27% to 0.47% in fresh plant. Samples of each
extract were made up at a concentration of 20 mg/ml for GC-MS analysis.
Table 5.4 The extract yields of supercritical carbon dioxide extractions
with ethanol entrainer
Material
Harvest
date
Extracting conditions
Weight of
extract
Yield
(dry
plant)
Yield
(fresh
plant)
Dry bracken
whole plant
90 g
May 2009
35 MPa, 50 oC, 4 hrs
EtOH 4 ml/min
0.82 g
0.90%
0.27%
Dry bracken
July 2009
35 MPa, 50 oC, 4 hrs
0.52 g
0.74%
0.35%
0.60 g
0.86%
0.47%
whole plant
70 g
Dry bracken
whole plant
70 g
EtOH 4 ml/min
October
2009
35 MPa, 50 oC, 4 hrs
EtOH 4 ml/min
Table 5.4 shows the extracted material and yield of each extraction with ethanol.
Compared to the yields of supercritical CO2 extract, the yields of extractions with
ethanol entrainer fluctuate within a relative limited range in both dry plant and fresh
plant. As previously stated, the seasonal variation of bracken extraction yield will be
discussed in more detail in a later chapter.
Figure 5.14 shows the gas chromatogram of each supercritical and EtOH extract with
the composition highlighted. The pattern of May 09 bracken sample is very similar to
that of crozier extract. Almost every peak that is present in crozier supercritical CO2
166
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
ethanol extract still exists in this more mature frond. Pterosin B remains at a high
concentration is this extract and afforded a major peak with the second greatest peak
area. Pterosin F peak in this extract is very weak compared to pterosin B, this may due
to the majority of this compound has been extracted out during previous supercritical
CO2 extraction. -sitosterol is still one of the major peaks in this extract; its quantity
was calculated as 509 g/g dry plant and 153 g/g fresh plant. As seen with the
crozier extract, n-hexadecanoic acid and phytol with related branched alcohols were
still found in this May 09 sample. Long chain primary alcohols were no longer
detectable in this extract possibly because they were fully extracted in previous
supercritical CO2 extraction.
The chromatogram of July 09 and October 09 samples are similar in composition.
Pterosin B and pterosin F were non-existent in these mature and over-mature fronds
and instead of -sitosterol, phytol and related branched unsaturated alcohols become
the major peaks in these extracts. The quantity of -sitosterol was calculated as 118
g/g dry plant and 55 g/g fresh plant in fully matured frond, and 181 g/g dry plant
and 100g/g fresh plant in senescing frond. C45 and C47 long chain primary alcohols
also present in this extract. However, their peaks are very weak and barely visible in
the chromatogram.
In conclusion, in supercritical CO2 with entrainer extracts, pterosin compounds also
exhibited a clear reduction with the seasons. Pterosin B and pterosin F disappeared in
the frond when it is growing towards its maturity. -sitosterol consecutively exists in
bracken frond in every life stage. Its quantity was all recorded and its variation
tendency will be discussed in later chapter.
167
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
509 bracken SCCO2+EtOH extract
, 29-Apr-2010 + 14:07:05
lzbco2+etoh509
19.92 20.56
100
n-hexadecanoic acid
Pterosin B
32.84
Scan EI+
TIC
1.72e8
-sitosterol
Phytol and
41.63 42.79 43.88 45.08
bracnched alcohols
%
3.51 4.80
13.60
23.79
16.78
25.77
27.23
31.11
0
lzbco2=etoh709
Bracken 7/09 SCCO2+EtOH extract
18.39
100
C45 & C47 long chain alcohols
Phytol and
42.83
bracnched alcohols
-sitosterol
%
3.05 4.79
13.60
9.50
47.97
21.93
18.45
9.49
46.84
39.11
45.24
Scan EI+
TIC
1.28e8
46.52 47.87
40.34 40.97
16.77
0
lzbscco2+etoh1009
Bracken 10/09 SCCO2+EtOH extract
18.34
100
C45 & C47 long chain
Phytol and
bracnched alcohols
-sitosterol
%
3.08 4.84
38.92
40.16 40.93
42.75
Scan EI+
TIC
alcohols 2.85e8
45.10
46.97
47.97
16.99
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 5.14: Gas chromatogram of bracken supercritical CO2 with EtOH extracts
168
40.51
45.51
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5.4 Investigation of pterosin precursor in lady fern (Athyrium
filix-femina) by using supercritical fluid chromatography
A sample of Lady Fern at crozier stage was harvested in April 2009 from Kilburn
White Horse (Sutton Bank, Grid reference SE 514 813) and was misidentified as
bracken fern. This lady fern sample was dried and pre-treated using the same method
as bracken fern. However, when the lady fern ethanol extract (extracted using method
in chapter 7.4.3) was analysed by GC-MS, an interesting peak was discovered and
was worthy of further studies in detail.
Figure 5.15 shows the chromatogram of lady fern ethanol extract. A peak with
retention time of 10.08 min was identified as 2,3-dihydrobenzofuran by matching its
mass spectra with NIST library standard spectra which has a very similar chemical
structure with indanone. Besides, the peak at 22.00 min was guessed to be methyl
lathodoratin which has a similar structure with pterosin B. Moreover, the
characteristic ions of methyl lathodoratin were found to be m/z 220, 205 which is very
similar to that of pterosin B (m/z 218, 203). Therefore, there is a conjecture that
although lady fern is not carcinogenic, there might be some form of ptaquiloside
related precursor present in lady fern and afforded this indanone-like metabolite. To
have a better understanding of the chemical similarity between lady fern and bracken
fern, supercritical fluid chromatography was employed due to its rapid analysis and
lower detection limit in ethanol extract compared to GC-MS.
Lady fern and bracken fern crozier (picked in May 2010) were extracted by ethanol
using the method described in section 7.4.3. The crude ethanol extracts were analysed
by supercritical fluid chromatograph directly using the method stated in 7.4.4. Figure
5.16 shows the supercritical chromatogram of bracken fern overlaid with lady fern
and with some mutual peaks and characteristic peak highlighted. The major peak in
bracken ethanol extract presented at 2.8 min and was postulated to be pterosin B since
169
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
it was the major peak in corresponding GC-MS sample. Two mutual peaks are also
observed at 5.0 min and 6.3 min.
170
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Expected indanone
R.T. 10.08min
2,3-dihydrobenzofuran
Structure of pterosin B
LZB-1
JHC23304JZL
100
TOF MS EI+
TIC
3.33e4
%
3.83 4.08 5.64
3.68
0
2.00
18.99
20.55 22.00
20.89
5.53
10.08
10.35 11.41 13.48
6.086.87 7.248.26
14.99
3.32
R.T. 22.00min
methyl
lathodoratin
O
O
19.53
22.82
OH
33.37
O
28.48
29.90 30.81 31.89
29.02
24.54
Time
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
22.00
24.00
26.00
Figure 5.15: Gas chromatogram of lady fern ethanol extract
171
28.00
30.00
32.00
34.00
36.00
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
At the moment, standard pterosin B is obtained by converting standard ptaquiloside
in alkaline and acidic conditions. Standard ptaquiloside itself was first extracted by
water from bracken fern, and then its isolation was accomplished by a series of
extreme complicated purification process. Since one of the major advantage of
preparative SFC is it combines rapid analysis with precise fractionation, thus it may
be able to make pterosin B isolation much more efficient and simple.
Possibly pterosin B
Mutual peaks
Figure 5.16: Supercritical chromatogram of bracken fern
and lady fern ethanol extracts
Since there is a high possibility that the peak at 2.8 min in bracken extract is pterosin
B, the fractionation technique was applied to isolate this peak so that it could be
used as pterosin B standard in future research. The fractionation was accomplished
by Waters Investigator SFC system and a fraction between 2.6 min to 3.0 min was
collected. A light green compound was obtained and was analysed by GC-MS for
identification.
172
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Figure 5.17 shows the gas chromatogram and mass spectra of the peak at 2.8 min in
bracken extract and the retention time comparison with C12-60 hydrocarbon
standard. The mass spectra of this compound indicates clearly that it is an alkane,
and after comparing its retention time with C12-60 hydrocarbon standard, the
number of carbon was confirmed to be 27 since the peak falls between hexacosane
and octacosane. Therefore, the major peak of bracken ethanol extract observed in
supercritical chromatogram was identified to be heptacosane rather than pterosin B.
Because the Waters Investigator SFC system was only available from one month
before this project finished, many further investigations on bracken and lady fern
mutual peaks and further pterosin B isolation were not accomplished because of
time constraints. Although the pterosin B searching in bracken fern ethanol extract
wasn’t achieved, the success of fractionation demonstrate the technique itself could
be a powerful technology for target compound purification and isolation. Further
research is required to reveal the chemical similarity between lady fern and bracken
fern, also further work in pterosin B fractionation by preparative SFC is needed as
well.
173
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
bracken SFC fraction 1
, 13-Sep-2010 + 12:27:42
lz-sfc fraction
Scan EI+
TIC
1.03e9
28.69
100
%
47.06
3.05
0
es c12-60 17-9-10
Scan EI+
TIC
1.58e9
29.98
100
26.17
31.68
21.75
16.43
13.35
28.15
24.05
19.21
33.29
36.22
38.86
9.95
42.07
%
37.96
3.01
0
Time
5.50
10.50
15.50
20.50
25.50
30.50
35.50
40.50
bracken SFC fraction 1
45.50
, 13-Sep-2010 + 12:27:42
lz-sfc fraction 3845 (28.692) Cm (3842:3851-3868:3884)
Scan EI+
1.22e8
57
100
71
43
%
85
55
42
44
69
58
72 83 86
99
109
113
127
141
155
0
29
49
69
89
109
129
149
169
169
183
189
197
211
209
225
229
239
254
249
268
269
281
289
337
309
329
Figure 5.17 Gas chromatogram and mass spectra of the fraction and gas chromatogram of C12-60 alkane standard
174
m/z
349
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
5.5 Conclusion and further remarks
Bracken crozier (May 2010 sample) was extracted by liquid CO2, supercritical CO2
and supercritical CO2 with ethanol entrainer sequentially. Pterosin B and pterosin F
were discovered in bracken liquid CO2 extract and were identified by comparing their
mass spectra with published standard spectra. Their existence possibly means that
these compounds are present in bracken crozier natually prior to the extraction
although the source of pterosin F in bracken frond still needs further investigation.
-sitosterol was also detected in crozier liquid CO2 extract, and its quantity was
calculated to be 1217 g/g dry plant and 324 g/g fresh plant by using stigmasterol as
a standard due to their structure similarity. C37 and C39 long chain primary alcohols
are also present in this extract and were identified by comparing their spectra and
calculated Kovats index with published standard data.
In bracken crozier supercritical CO2 extract, all these peaks listed are still present with
some new peaks appearing due to the change of extracting conditions. Phytol and
some related branched unsaturated alcohols were discovered along with more long
chain primary alcohols. C37-C47 odd number carbon alcohols were found and this
demonstrates that the extracting ability of scCO2 is significantly higher and it is now
able to extract compounds with much higher molecular weight than liquid CO2.
-sitosterol was found to be 754 g/g dry plant and 201 g/g fresh plant in this
extract.
Pterosin F and all long chain alcohols were absent in bracken supercritical CO2 with
ethanol entrainer extract. This may due to these compounds have been completely
extracted in previous extractions. However, major peaks such as pterosin B and
-sitosterol were still present in this extract. -sitosterol has a concentration of 702
g/g dry plant and 187 g/g fresh plant in this extract.
175
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
Bracken fronds in three more mature life stages were also extracted by supercritical
CO2 and supercritical CO2 with ethanol entrainer. The reduction of pterosin
compounds with seasons is very obvious. Pterosin B and pterosin F are present in
May 09 frond as major peaks. When it gets fully matured in July, pterosin B is barely
detectable and pterosin F is totally non-existent. In October senescing bracken, both
pterosin B and pterosin F have disappeared. The changing tendency of pterosins in
bracken was discussed in detail in a later chapter.
-sitosterol presents consistently in bracken of every life stage. It was found to be
1074 g/g dry plant and 318 g/g fresh plant in May 09 sample, 660 g/g dry plant
and 309 g/g fresh plant in July 09 sample and 1132 g/g dry plant and 625 g/g
fresh plant in over-matured October sample. Besides this, C37-C47 long chain
primary alcohols also present persistently in bracken supercritical CO2 extract of
every life stage. The seasonal variation of -sitosterol will be discussed in later
chapters.
In bracken supercritical CO2 with ethanol entrainer extracts, pterosin B is present with
a high peak area and is the major peak in May pre-mature frond. However, compared
with pterosin B, pterosin F peak is very weak in this extract. As season changes, both
pterosins disappear as the bracken grows towards its maturity. A further discussion on
seasonal variation of pterosins will be brought out in a later chapter.
In contrast, phytol and related branched alcohols become noticeable in fully matured
bracken and afford major peaks in the July and October bracken extract. Two largest
long chain alcohols (C45 and C47) were also found in July and October bracken.
-sitosterol exists persistently in bracken supercritical CO2 with entrainer extracts. It
was found to be 509 g/g dry plant and 153 g/g fresh plant in May sample, 118 g/g
dry plant and 55 g/g fresh plant in July frond, and 181 g/g dry plant and 100g/g
fresh plant in senescing October frond. Its seasonal variation trend will be discussed
in detail in later chapter.
176
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
In bracken crozier sequential carbon dioxide extractions, the raw material has been
extracted by liquid CO2, supercritical CO2 and supercritical CO2 with ethanol
entrainer continuously for 14 hours. Pterosin B was found in the extracts obtained in
every step and pterosin F was found in the first two extracts. The reason for their
consistent existence may be their insufficient solubility in liquid CO2 and supercritical
CO2. Since pterosin B and F do not have enough solubility in liquid CO2, after 6 hours
liquid CO2 extraction, there was still some residual pterosins left in the material.
Therefore, in the subsequent supercritical CO2 extraction, more pterosins were
extracted due to their enhanced solubility in supercritical CO2. Similarly, pterosin B
was still present in supercritical CO2 with entrainer extract due to its insufficient
solubility in supercritical CO2, when pterosin F is completely extracted in the second
step. Therefore, it is clear that compared to liquid CO2, supercritical CO2 is preferable
to give a maximum extracting ability and achieve higher yield despite a reduction in
selectivity. When considering a full scan of plant chemical composition, supercritical
CO2 extraction would be a better choice than liquid CO2 extraction.
Finally, supercritical fluid chromatography was applied to isolate pterosin B and to
study the chemical similarity between lady fern and bracken fern. Two mutual peaks
were found in lady fern and bracken fern ethanol extracts. Meanwhile, a characteristic
peak of bracken ethanol extract was successfully fractionated by semi-preparative
SFC system. However, this fraction was identified as heptacosane by GC-MS. Due to
the time constraints, further isolation and identification of mutual peaks and pterosin
B was not accomplished. Apparently, if time allows, more study will be carried out to
achieve this incomplete research.
177
Chapter 5
Liquid and supercritical carbon dioxide extractions of bracken
178
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
Seasonal variation of pterosins and
other major compounds in bracken
Chapter 6
179
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
6.1 Seasonal variation in ptaquiloside derivatives pterosins and other
major compounds
Bracken fern (Pteridium aquilinum) is a widely distributed species that dominates the
vegetation of large areas in many UK upland regions. Bracken is a perennial fern, it’s
also the only higher plant known to cause cancer naturally in animals. The content of
the major toxin and carcinogen ptaquiloside varies widely throughout the seasons. It
was reported that the carcinogenicity of young bracken shoots was shown to be
stronger than that of mature plants.6 During bracken’s life cycle, the crozier (Figure
6.1) contains the highest concentration of ptaquiloside. This is due to its
self-protecting mechanism against herbivores.90
Figure 6.1: The crozier step in
bracken vegetation process
It is reported that the concentration of ptaquiloside in the frond decreases over time.91
When the crozier start to stretch and generate the first pair of leaves (Figure 6.2), the
ptaquiloside is still present at fairly high concentration. As the growth cycle
180
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
progresses, ptaquiloside starts to reduce dramatically and end up with a low level in
mature (Figure 6.3) and dying plants (Figure 6.4).
Figure 6.2: 2nd step in
bracken’s life cycle-with
first pair of leaves
Figure 6.3: 3rd step in
bracken’s life
cycle-mature plant
Figure 6.4: 4th step in
bracken’s life cycle-dying
bracken
Therefore, samples in different bracken life cycle steps are picked from the same site
(Kilburn White Horse, Sutton Bank, Grid reference SE 514 813) in different months
of a year. The first crozier sample was picked in 17th April 2009 and 19th May 2010,
181
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
followed by the 2nd step sample picked in 21st May 2009, 3rd sample picked in 27th
July 2009 and 4th sample picked in 12th October 2009. Unfortunately, the sample
picked in April 2009 was misidentified and found to be lady fern rather than bracken
fern, all the data and information for bracken crozier was obtained from May 2010
sample. In this chapter, seasonal variation of pterosins and some other compounds in
bracken will be examined and will demonstrate a clear trend of concentration of
ptaquiloside related compounds.
6.2 Water content variation of bracken
Water content of bracken fern was examined in samples at different life cycle steps
from April to October 2009 (Figure 6.5). The highest water content was found to be
73.40% by total weight in bracken crozier sample (05/2010). When the plant has the
first pair of leaves, the water content dropped slightly to 70.40% but this trend
accelerated as the plant becomes more mature. It decreased sharply by almost 20%
within 2 months and reached a figure of 53.20% in July 2009. Finally, the decrease in
water content continued until it reached 44.80% in over-matured frond.
Water content (%)
80.00%
73.40%
70.40%
70.00%
53.20%
60.00%
44.80%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
05/2010
05/2009
07/2009
10/2009
Month
Figure 6.5: Seasonal variation in water content of
bracken fern
182
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
6.3 Variation in total extract yield of bracken over a life cycle
Extract yields of supercritical CO2 extractions with and without ethanol entrainer are
listed below (Table 6.1). Dry bracken samples were extracted using supercritical CO2
at 350 MPa and 50 oC for 4hrs on the first day, and were then extracted with the
addition of ethanol as a co-solvent at 10% of the CO2 flow under the same conditions
on the following day. Two yields were recorded respectively and a total yield was
made up by adding these two together. Figure 6.6 shows the trend in yield change
throughout the year.
Table 6.1 Extract yield of supercritical carbon dioxide extractions
1
Material
Extracting conditions
Yield
Total yield
Dry bracken 05/10
35 MPa, 50 oC, 4 hrs
1.42%
2.19%
Dry bracken 05/10
whole plant 90g
35 MPa, 50 oC, 4 hrs +
EtOH
0.77%
Dry bracken 05/09
whole plant 90g
35 MPa, 50 oC, 4 hrs
1.31%
Dry bracken 05/09
whole plant 90g
35 MPa, 50 oC, 4 hrs +
EtOH
0.90%
Dry bracken 07/09
whole plant 70g
35 MPa, 50 oC, 4 hrs
1.63%
Dry bracken 07/09
whole plant 70g
35 MPa, 50 oC, 4 hrs +
EtOH
0.74%
Dry bracken 10/09
whole plant 70g
35 MPa, 50 oC, 4 hrs
1.84%
Dry bracken 10/09
whole plant 70g
35 MPa, 50 oC, 4 hrs +
EtOH
0.86%
whole plant 90g
2
3
4
2.22%
2.37%
2.70%
The total yield was found to vary from 2.19% to 2.70%. The highest supercritical CO2
extract yield and total yield were both obtained in October 2009, with a figure of 1.84%
183
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
and 2.70% respectively. However, the highest CO2 with entrainer extract yield was
obtained in May 2009 with a level of 0.90%. The supercritical CO2 and total yield
from dried bracken increase over time, until they reach a peak in the senescing plant.
However, the yields of CO2 with entrainer remained stable over the season, with some
fluctuation within very limited range.
3.00%
Extracting yield (%)
2.50%
2.00%
1.50%
1.84%
1.42%
1.31%
1.63%
0.77%
0.90%
0.74%
0.86%
05/2010
05/2009
07/2009
10/2009
1.00%
0.50%
0.00%
Month
supercritical carbon dioxide+EtOH extacting yield
supercritical carbon dioxide extracting yield
Figure 6.6 Seasonal variation in supercritical extraction
yield and total yield of dry plant
Taking into account the water content, the total yield and individual extract yields
were also examined in fresh plant material. Figure 6.7 shows the seasonal variation of
total yield and different supercritical extract yield from fresh plant material. Overall,
the changing tendency of fresh plant extracting yield is exact the same with dry plant.
The supercritical CO2 yield increases remarkably over the season and reach the
highest amount in October at 1.02%. In contrast, the increasing rate of CO2 with
entrainer yield is much milder. The yield is at its lowest point in the crozier extraction
at 0.20% and then rises until it reaches a peak in October. The trend of increase of
184
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
fresh plant material is even more significant than dry plant. The differences between
yields are more obvious when water content is taken into account.
1.60%
Extracting yield (%)
1.40%
1.20%
1.02%
1.00%
0.80%
0.76%
0.60%
0.40%
0.20%
0.38%
0.39%
0.20%
0.27%
0.35%
05/2010
05/2009
07/2009
0.47%
0.00%
10/2009
Month
supercritical carbon dioxide+EtOH extracting yield
supercritical carbon dioxide extracting yield
Figure 6.7 Seasonal variation in supercritical extract yield
and total yield of fresh bracken
6.4 Seasonal variation of pterosins in bracken
As mentioned before, pterosin B and pterosins F are naturally occurring pterosins in
bracken supercritical CO2 extract. It was reported before that when the frond
approaches towards its maturity, the tender tissues of the crozier become harder and
less attractive for herbivore. Therefore compounds that protect the crozier such as
ptaquiloside and pterosins will become no longer necessary in the organism at a same
high concentration. They are either metabolized or transferred to water or soil etc.2
185
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
5/09
bracken
SCCO2
5/10
crozier
scco2extract
extract
Pterosin B
Pterosin F
lzbsc510k
20.40
100
, 26-Apr-2010 + 13:01:53
-sitosterol
Scan EI+
TIC
3.50e8
32.64
19.94
%
21.56
32.77
31.94
0
lzbscco2-509
5/09
100
bracken scco2
extract
40.70
39.27
38.07
42.47
41.82
35.48 36.87
43.55
44.74
46.05 47.63
Scan EI+
TIC
3.94e8
-sitosterol
20.57
Pterosin B
Pterosin F
%
32.86
39.53 41.01
19.93
21.73
32.16
27.26
0
lzbscco2-709
32.99 34.37
38.28
40.99
32.86
13.63
45.26
45.02
34.39
32.41
20.57
27.26
37.64
29.37
0
lzbscco2-1009
10/09 bracken scco2 extract
-sitosterol
100
40.93
42.76
45.10
44.47
Scan EI+
TIC
3.02e8
47.77
39.45
38.99
32.79
%
18.40 19.87
Scan EI+
TIC
47.841.57e8
38.85
%
18.46
42.89
42.59
39.50
21.73
3.03
47.75
37.07
-sitosterol
7/09 bracken scco2
extract
100
42.88 44.58 45.29 45.97
31.32
21.68
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 6.8 Original GC traces of seasonal bracken supercritical CO2 extract
186
40.51
45.51
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
Figure 6.8 shows the original GC traces of bracken supercritical CO2 extract samples.
As can see from these traces, in early bracken samples (crozier and May 09 samples),
pterosin B is present at a very high concentration as the peak is the major peak in
these two traces. Meanwhile, pterosin F is also present at a relatively high
concentration and its peak is the third highest peak in the chromatogram. After that, in
July sample, pterosin B and pterosin F decreased dramatically and are only present at
very low concentration. Finally, as the plant enters senescence in October, pterosin B
and pterosin F scarcely exist and are even not able to be detected by GC-MS. It is
apparent from these results that young fronds contains the highest concentration of
pterosins, then after the frond advance towards mature, pterosins will reduce sharply
and finally disappear in fully matured plant. This changing tendency strongly
corroborates previews studies and makes the result even more convincing.
Due to lack of pterosin standards, the quantification of pterosins could not be carried
out. Therefore, to confirm the GC analysis, a direct comparison of the relative level of
pterosins and octadecane internal standard in peak height/area was done. Figure 6.9
shows the GC traces of different bracken samples with octadecane as internal standard
at a consistent concentration of 1mg/ml. Since the concentration of octadecane is
identical in every single sample, comparing the relative peak height/area between
octadecane and pterosins, the trend of seasonal variation will be even more obvious.
As can see from the diagram, in crozier sample (trace 1, 5/10), the level of octadecane
and pterosin B peaks are almost the same with octadecane slightly higher than
pterosin B. Meanwhile, pterosin F peak is about half height of octadecane. Thereafter,
in the second sample, peak height of PtB falls to about 2/3 of octadecane and PtF only
present at a 1/3 height of octadecane. As the plant becomes more mature, PtB and PtF
are only present at very low concentration and barely exist in senescing plant since
their peaks are hardly detected by GC. The direction of change based on the relative
level comparison of pterosins and octadecane again confirmed the previous
conclusions.
187
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
bracken
scco2 scCO
extract with
5/105/10
bracken
withstandard
std
, 24-Feb-2011 + 17:37:54
2
lz-b510final
Octadecane
1mg/ml
100
33.65
18.45
%
PtF
21.21
20.78
13.94
PtB
38.45
27.89
22.34
39.06 40.54
42.31
44.58
40.54
42.32
44.56
46.85
44.61
47.46
44.60
47.37
32.95
0
lz-b509final
5/09 bracken scCO2+std
100
Octadecane
1mg/ml
Scan EI+
TIC
47.35 2.20e8
39.08
18.46
Scan EI+
TIC
1.63e8
38.47
33.64
20.49 21.20
%
27.89
22.53
0
lz-b709final
7/09
bracken
scCO2+std
100
Octadecane
1mg/ml
32.95
40.54
39.09
18.47
42.34
Scan EI+
TIC
1.66e8
%
0
lz-1009final
10/09 bracken scCO2+std
100
Octadecane
1mg/ml
38.80
18.46
39.97 40.55
42.32
Scan EI+
TIC
1.57e8
%
3.02
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
40.51
Figure 6.9 GC traces of seasonal bracken supercritical CO2 extract with octadecane standard
188
45.51
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
Apart from the seasonal variation of pterosins, another interesting phenomenon can be
observed from the two figures of GC traces. In figure 6.8, when the samples were
freshly extracted, pterosins especially pterosin B used to exist as the highest peak in
the extract. Particularly in the second sample, the peak height of pterosin B is
approximately twice that of -sitosterol. That indicated in the early stages of growing,
pterosins are always the major compound in the frond. However, when the very same
extracts were analysed again nine months later with internal standard by GC at the
same initial concentration, the situation changed greatly. The peak height of pterosin
peaks falls by 1/3 or even half and are much shorter than -sitosterol. That means
pterosins have degraded drastically during storage. It is well known that ptaquiloside
has a very unstable nature. It can readily decompose in both acidic and alkaline
conditions. It is generally acknowledged that pterosin B is much more stable
compared to ptaquiloside and it is the main final metabolite of ptaquiloside
decomposition. This discovery of pterosin B degradation draws attention to stability
as an issue, and a further investigation is needed to ascertain the mechanism of this
degradation.
6.5 Seasonal variation of -sitosterol
-sitosterol is one of the main compounds in bracken supercritical CO2 extract. It
appears in bracken samples of every growth stage at a high concentration. To the best
of my knowledge, the seasonal variation of sitosterol in bracken has never been
researched before. Since -sitosterol possess some impressive biological properties
such as blood cholesterol lowering effect, it is necessary to have some further
investigation on seasonal variation of it in bracken. Because of the high content of
-sitosterol, bracken could be a potential source of this high value chemical.
Therefore quantification of -sitosterol in bracken is significant.
189
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
The quantification of -sitosterol was carried out with the application of stigmasterol
as a standard. Standard solutions of stigmasterol in different concentrations were
made up and analysed by GC. The absolute areas were recorded and a standard curve
was determined using these area data. Figure 6.10 shows the standard curve of
stigmasterol and the regressive equation. A correlative coefficient of 0.994 was
Area of peak (pA*s)
Millions
obtained from this curve.
18
y = 6E+06x - 1E+06
R² = 0.9946
16
14
12
10
8
6
4
2
0
0
0.5
1
1.5
2
2.5
3
3.5
Concentration of stigmasterol (mg/ml)
Figure 6.10 The standard curve of stigmasterol
for -sitosterol quantification
-sitosterol content of bracken samples extracted by supercritical CO2 and
supercritical CO2 with ethanol entrainer were examined respectively. The total
-sitosterol content in bracken dry plant was also determined by adding these two
figures together. Figure 6.10 shows the -sitosterol content in every extract and the
variation tendency throughout the year.
The pattern of change in concentration of -sitosterol is not similar to that of any
previous compounds. The greatest total concentration of -sitosterol, 2176 g/g dry
plant was obtained in the first crozier sample. Then the -sitosterol content declined
190
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
significantly to 1583 g/g in second stage sample and this trend continued and
reached the lowest point in June with a figure of 778 g/g. However, the content of
-sitosterol finally rebounds in over-mature plant with a level of 1312 g/g dry plant.
It is clear that the variation in level of -sitosterol is similar to that of pterosins before
the plant gets over-mature. The abundance of -sitosterol is higher in the early stages
of growing, and then reduces gradually over time. Unlike the pterosins, -sitosterol
finally rises up in dying plant and the mechanism behind this needs further
Abundance (g/g dry plant)
investigation.
2500
2000
592.9
1500
1000
508.5
180.6
1583.3
118.4
1074.2
500
1131.6
660.15
0
05/2010
05/2009
07/2009
10/2009
Month
supercritical carbon dioxide only
supercritical carbon dioxide+EtOH
Figure 6.11 Seasonal variation of -sitosterol
in bracken dry plant
Quantitative variation of -sitosterol in fresh plant material was also examined. Figure
6.12 shows the seasonal variation of -sitosterol in fresh bracken. The downward
trend in earlier samples remains the same with dry plant. 579 g/g -sitosterol was
found in crozier fresh sample, at the same time 469g/g and 364g/g was found in
May and July bracken respectively. Surprisingly, when water content is taken into
account, the greatest total concentration of -sitosterol was obtained in October
sample rather than the crozier. A total of 724 g/g -sitosterol was found in senescing
191
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
plants and this sample also contains the highest concentration of -sitosterol in CO2
with entrainer extract. In summary the abundance of -sitosterol in fresh plant is high
in the crozier then reduces gradually over time, later in the growing season the
-sitosterol content increases dramatically and shows the highest abundance in
senescing plants. As mentioned before, the mechanism behind this rebound needs
Abundance (g/g dry plant)
further investigation.
800
99.69
700
600
500
157.75
150.52
400
55.41
300
200
421.15
317.97
624.65
308.96
100
0
05/2010
05/2009
07/2009
10/2009
Month
supercritical carbon dioxide+EtOH
supercritical carbon dioxide only
Figure 6.12 Seasonal variation of
-sitosterol in bracken fresh plant
6.6 Pterosins variation caused by geographical difference
In order to understand the influence of geographical location on bracken pterosins
content, additional bracken samples was picked from Penrallt in North Wales in May
2010. Since bracken crozier contains the highest concentration of pterosins, these
samples picked at the same time of the year should indicate the impact of the location
difference at a time when the pterosins are at their highest. The Welsh bracken sample
192
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
was extracted using supercritical CO2 under identical conditions to the Yorkshire
samples and analysed by GC-MS.
Figure 6.13 shows the chromatogram of bracken extract from the different locations.
Due to the lack of pterosin standards, quantification was not accomplished. However,
in the integrated chromatogram, -sitosterol has a very similar absolute area in both
samples (within 1.17% difference) and since the two samples are prepared at the same
initial concentration (20 mg/ml) the levels can be seen relative to the concentration of
-sitosterol.
193
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
bracken SCCO2 extract 5/10 wales
5/10 Kilburn bracken sample
lzbsc510k
, 19-May-2010 + 15:52:46
20.40;11242999
100
Area
Scan EI+
TIC
3.50e8
32.64;11332011
-sitosterol
PtB
PtF
40.70
42.47
39.27
5356378
4518786
3718789
19.94
3818783
%
19.76
2371005
21.56
1617370
31.94
906391
38.07
1676867
32.77
938131
44.74
2801725
0
lzbsc510w
5/10 Wales bracken sample
Scan EI+
TIC
2.95e8
32.65;11199506
100
-sitosterol
Area
%
40.72
4005814 42.52
39.29
2905863
2225381
20.40
2016765
31.96
766201
44.79
3145267
38.09
951796
32.80
605331
0
Time
5.51
10.51
15.51
20.51
25.51
30.51
35.51
Figure 6.13 Comparison of Kilburn and Wales bracken supercritical CO2 extract
194
40.51
45.51
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
As can be seen from the first trace of Kilburn sample, pterosin B is present at a very
high concentration, approximately the same as that of -sitosterol. Meanwhile,
pterosin F also exists at a fairly high concentration and its peak area was slightly more
than one third of pterosin B. It is clear that in this sample, pterosins, particularly
pterosin B, are the major compounds in this extract along with -sitosterol. On the
other hand, in Welsh bracken sample is completely different. Pterosin B peak has less
than one-fifth area of -sitosterol, and pterosin F has less than one-tenth area of
-sitosterol. The quantity difference of pterosins in Kilburn and Welsh bracken
samples are very obvious. Welsh bracken contains less than 40% of total pterosins
compared to Kilburn bracken. This distinct difference may be due to different pH
value of the soil in Kilburn and Wales.
Kilburn
Penrallt
Figure 6.14 Map of soil pH in the UK92
195
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
According to the map of soil pH in the UK supplied by National Soil Resources
Institute,92 the soil pH value around North Yorkshire Moore is quite different from
Wales. In most land around York and North Yorkshire Moore, the top 15cm of soil is
neutral or alkaline (pH 6.6-7.5) and the White Horse at Kilburn where the samples
were collected is immediately below a limestone cliff. Conversely, in Wales, most of
the area has strong to moderate acid soil (pH below 4.5-6.5) and on the Lleyn
peninsula lime is regularly added to arable land to increase the naturally low pH. In
consideration of the ptaquiloside decomposition pathway, pH value variation between
these two sites may be the most likely explanation. Since the ptaquiloside will more
rapidly degrade in alkaline conditions than in acid, their lifetime in Kilburn bracken
will be shorter than in the Welsh sample. According to Natori group,87 the half-live of
ptaquiloside in aqueous solution is 5 days for pH 4.0 and 23 min for pH 10.0. This
means alkaline condition is favorable to ptaquiloside decomposition. Thus the
alkaline soil of Kilburn may accelerate the ptaquiloside degradation and result in the
increasing in the final metabolites pterosin B and pterosin F. The pterosins content in
the supercritical CO2 extract may come from three sources: first, pterosins that
originally appear in the frond; second, pterosins degraded from ptaquiloside prior to
extraction and third, pterosins generated from ptaquiloside degradation during the
extraction (because of the acidic extracting conditions). In this case, apparently
because of the basic soil, in Kilburn bracken, much more ptaquiloside is converted to
pterosins thus it is more likely that the second scenario is correct and can explain why
Kilburn bracken contains higher levels of pterosins compared to the Welsh sample.
However, because standards were unavailable, quantification of pterosins from each
source in both bracken samples cannot be accomplished. Therefore, further work is
required to quantify exactly the level of pterosins from each source when standards
are available.
196
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
6.7 Conclusion and further remarks
Seasonal variation of some major compounds of bracken supercritical CO2 extract
was highlighted in this work. A relative-level comparison between pterosins and
internal standard was applied to mitigate the lack of standards. Pterosins variation
largely reflected previous studies. Pterosins are present at very high concentration
early in the growth cycle but reduce dramatically when the plant approaches
senescence. Barely no pterosins can be detected from over-mature samples, and the
reason behind this variation may be a self-protecting mechanism. Degradation in
pterosins was also found during this work. Pterosins are significantly reduced when
repeat analysis was carried out on samples after nine months. Further work is required
to find out the mechanism behind this decomposition.
Seasonal variation of another important compound -sitosterol is also demonstrated in
this work. Quantitative results from each bracken sample were obtained and the
changing trend is obvious. Bracken crozier contains -sitosterol at 579 g/g in the
fresh plant material and this declined gradually until it reached the lowest point at 364
g/g in fully mature bracken. Later in the growth cycle concentration increased and
finally reached 724 g/g in over-mature fronds.
It is also discovered that geographical difference can affect the pterosins content.
Bracken samples from Kilburn and Wales picked at the same time period of the year
were extracted and compared. The result indicated Kilburn samples contains almost
three times as much pterosins as the Welsh sample. This could due to the pH value
variation between these two places. Generally speaking, the soil pH value is lower in
the area of harvest in Wales than at Kilburn in NorthYorkshire. Since ptaquiloside is
more unstable in alkaline condition than in acidic condition, more ptaquiloside in
Kilburn bracken may be converted to pterosins prior to harvest and extraction.
Therefore, higher levels of pterosins were obtained from Kilburn bracken than Welsh
197
Chapter 6
Seasonal variation of pterosins and other major compounds in bracken
bracken after the extraction. This work indicates that geographic location indeed has a
remarkable influence in plant chemical composition.
198
Chapter 7
Materials, methods and experimental procedures
Materials, methods and experimental
procedures
Chapter 7
199
Chapter 7
Materials, methods and experimental procedures
7.1 Materials and reagents
Liquid carbon dioxide (99.99%) and bulk liquid nitrogen were purchased from BOC
Ltd. All analytical and HPLC grade solvents, ethanol, methanol, acetone,
dichloromethane, toluene, ethyl acetate, diethyl ether, and acetonitrile were obtained
from Fisher Scientific.
Phenolic acid standards, gallic acid, (+)-catechin, (-)-epicatechin and chlorogenic acid,
Folin-Ciocalteu’s phenol reagent 2N, triterpenoid standards such as ursolic acid 90%,
oleanolic acid 97% and lupeol 98% were all purchased from Sigma-Aldrich.
For retention time calculations alkane standard C12-C60 was purchased from
Sigma-Aldrich and the C21-C40 standard solution was purchased from Fluka.
Samples of Calluna vulgaris were collected from the North Yorkshire Moor National
Park, North Yorkshire regularly from October 2007 to November 2008 with the
consent of North York Moor National Park Authority. Samples of Pteridium
aquilinum were collected from Kilburn, North Yorkshire (Grid reference SE 514 813)
in April, May, July, October 2009 and May 2010 and from Penrallt, Gwynedd (Grid
reference SH 370 355) in May 2010.
7.2 Harvesting and milling process of heather and bracken
7.2.1 Harvesting and preparation of heather material
The heather was cut with garden scissors at 5 cm above the ground and was bagged in
paper sacks for transportation. The raw material was air-dried in a fume hood at room
temperature for 5 days and then ground using a Glen Creston Ltd. hammer mill fitted
200
Chapter 7
Materials, methods and experimental procedures
with a 2.5 mm screen. The processed material was kept in glass sample bottles and
stored at -18 oC prior to extraction.
Figure 7.1: Milled whole heather
Flowering Calluna vulgaris was collected in August 2008 and the flowers were
separated manually. The separated flowers were kept in glass sample bottles and
stored at -18 oC prior to extraction.
Figure 7.2: Flowering heather
Figure 7.3: Separated heather flowers
201
Chapter 7
Materials, methods and experimental procedures
7.2.2 Harvesting and preparation of bracken
Previous research has indicated that the very young frond of bracken, known as the
crozier, contains the highest level of carcinogen compared with other life cycle steps.
Samples were collected in April, May, July and October from Kilburn, North
Yorkshire (Grid reference SE 514 813) to obtain bracken at different growth stages so
that the change in ptaquiloside and pterosins could be determined throughout the year.
The bracken was cut by hand at a height of less than 2cm above ground level and was
bagged in paper sacks for transportation. The raw material was air-dried in a fume
hood at room temperature for 2 weeks and then milled using a Glen Creston Ltd.
hammer mill fitted with a 2.5 mm screen. The processed material was kept in glass
sample bottles and stored at -18 oC prior to extraction. Because of the toxicity of the
plant material protective clothing and gloves were worn at all stages of harvest and
processing.
7.2.3 Water content of heather
Whole unmilled plant was dipped into liquid nitrogen then transferred to a one litre
freeze drying vessel and placed in an ice chest at -30 oC for several hours. Thereafter
freeze drying was operated overnight. A Heto DW1 lyophiliser manufacture by
Heto-Holten A/S was used and the temperature was kept at -100 oC. After drying, the
material was then weighed again in order to calculate the water loss. The water
content was calculated using the following equation:
% Water content = [(Initial mass – Mass after drying) / Initial mass] × 100
202
Chapter 7
Materials, methods and experimental procedures
7.3 Extraction and fractionation of heather
7.3.1 Soxhlet extractions of heather
20 g of ground heather was placed in a Soxhlet thimble and placed into a Soxhlet
extractor. A 1 litre round bottomed flask was connected to the Soxhlet extractor
together with a double-wall condenser. 400 ml hexane (or ethanol) was used as
extraction solvent during the process. The solution was heated under reflux for 6
hours. The reaction solution was cooled then filtered and the solvent was evaporated
using a Buchi R-200 rotary evaporator. The crude extract was then weighed and the
yield was calculated.
7.3.2 Supercritical carbon dioxide extraction of heather
Supercritical CO2 extraction was carried out by using a Thar Supercritical Fluid
Extraction 500 system (Thar technologies). Extraction was carried in a 500 ml
extractor at 35 MPa and 50 oC for 4hrs with a CO2 flow rate of 40 g/min. After the
extraction was complete the extract was washed out of the separator using
dichloromethane and the solvent was removed by rotary evaporator to obtain the yield.
This was a necessary step as the extract yield was very low.
The supercritical CO2 extractions of heather with an ethanol entrainer was carried out
under the same condition (35 MPa, 50 oC) with ethanol added at a flow rate of 4
ml/min (10% of CO2 flow rate). The ethanol solution was released regularly from the
separator and combined at the end of the extraction before being evaporated to obtain
the yield.
203
Chapter 7
Materials, methods and experimental procedures
7.3.3 Fractionation of crude heather extract using column chromatography
200 mg of crude extract was dissolved in 10 ml dichloromethane in a round bottom
flask with a small amount of Kieselgel silica gel. The dichloromethane was then
evaporated using a rotary evaporator in order to coat the extract onto the silica gel.
The silica gel and sample mixture was then transferred on top of a 20 cm x 1 cm
slurry packed column of Kieselgel 60 silica gel. The upper surface of the column was
covered with an approximately 2 mm layer of sand. The column was then eluted with
hexane, hexane/diethyl ether (90/10, v/v), hexane/diethyl ether (4/1, v/v),
hexane/diethyl ether (1/1, v/v), hexane/diethyl ether (1/4, v/v) and diethyl ether
continuously. The eluent was collected and analysed by TLC. Fractions are combined
based on their spot positions revealed by TLC and were analysed by GC and GC-MS
later.
7.3.4 Thin layer chromatography (TLC) of crude heather extract and fractions
obtained after column chromatography
20mg of crude extract was dissolved in 1ml diethyl ether. A spot of the extract
solution was placed approximately 1 cm from the bottom of the TLC plate using a
micropipette. Using the method of Yamashiro,93 the plate was developed using
hexane and toluene sequentially to full height and then a solvent mixture
(hexane/diethyl ether/acetic acid, 70/30/1 v/v) was allowed to reach the half height of
the plate. The plate was dried in air for several minutes and then placed under UV
light to identify the position of the components. Alternatively the plate was dipped
into a staining solution (10% w/w phosphomolybdic acid in ethanol), and dried in an
oven at 120 oC for 2 minutes.
For the eluent obtained following column chromatography, the fractions were spotted
onto the plate without any diluting or concentrating. The fractions were combined
based on their spot positions revealed by TLC and were analysed by GC and GC-MS.
204
Chapter 7
Materials, methods and experimental procedures
7.3.5 Anti-microbial assays using heather extracts
7.3.5.1 Sample preparation for paper disc antimicrobial test (qualitative test)
Ground heather (200 mg) was added to 2 ml ethanol/water (80:20) and stirred for 2
hours at room temperature. The mixture was centrifuged at 3000 rpm for 15 min and
the supernatant was decanted. Residual plant material was extracted again under
identical conditions and supernatants were combined. Solvent was removed using a
rotary evaporator.
7.3.5.2 Preparation for multi-well plate antimicrobial test (quantitative test)
Ground heather (1 g) was added to 10 ml ethanol/water (80:20) and stirred for 4 hours
at room temperature. The supernatant was filtered and collected, and then the solvent
was removed by a rotary evaporator. After the yield calculated, the solid residue was
dissolved in 2% ethanol water solution to give a concentration of 2 mg/ml. The
solution was sterilised using a Corning® syringe filter before adding to the 96-well
cell culture plate.
7.3.6 UV analysis of heather total phenolic content
Analysis of total phenolic content was carried out using Folin-Ciocalteu reagent and
absorbance measured at 765 nm using a Jasco V-550 UV/Vis spectrophotometer
according to the method of Andrew L. Waterhouse.94 A calibration curve was
produced using 0, 20, 40, 60, 80, 100 and 120 g/ml of gallic acid made up in
methanol and used as standards. 10l of each standard solution was added to a 10 ml
volumetric flask containing 1 ml Folin-Ciocalteu reagent and 5 ml 7.5% Na2CO3
water solution. Deionised water was then added to each volumetric flask to make up
the solutions to 10 ml. The solutions were incubated for 1 hour in the dark after
mixing well and then the absorption measured at 765 nm. Linear calibrations were
205
Chapter 7
Materials, methods and experimental procedures
produced by plotting absorption against concentration. Using the samples prepared in
7.3.5.2, 10 l of each heather ethanol/water extract was added to a 10 ml volumetric
flask containing 1 ml Folin-Ciocalteu reagent and 5 ml 7.5% Na2CO3 water solution.
Deionised water was then added to each volumetric flask to make up the solutions to
10 ml. The solutions were incubated for 1 hour in the dark after mixing well and then
the absorption measured at 765 nm. The total phenolic concentration of each heather
extract solution sample was then determined based on the calibration curve and
expressed in mg gallic acid equivalents (GAE) g-1.
7.3.7 UV analysis of heather total triterpenoid content
Analysis of total triterpenoid content was carried out using Vanillin-perchloric acid as
coloring reagent and absorbance measured at 550 nm using a Jasco V-550 UV/Vis
spectrophotometer according to the method of W. Wei.95 10 mg oleanolic acid was
dissolved in 50 ml methanol, then 0.00, 0.10, 0.20, 0.30, 0.40, 0.50, 0.60, 0.70 ml of
standard solution was transferred to 8 testing tubes and the methanol was evaporated
in a water bath. 0.3 ml 5% vanillin acetic acid solution and 1 ml perchloric acid were
added to each testing tube before heating these testing tubes in water bath at 70 oC for
25 min. The solution was cooled in ice water bath after heating and then 10.0 ml
acetic acid was added to each testing tube. Absorbance of each solution was then
measured at 550 nm and recorded to create a standard curve. Using the samples
prepared in 7.3.1 and 7.3.2, 10 mg of each heather supercritical CO2 extract (or
hexane extract) were dissolved in 1ml dichloromethane in testing tubes. The solvent
was then evaporated in a water bath. Same 0.3 ml 5% vanillin acetic acid solution and
1ml perchloric acid were added to each extract before heating these testing tubes in
water bath at 70 oC for 25 min. The solutions were cooled in ice water bath after
heating and then 10.0ml acetic acid was added to each testing tube. Absorbance of
each extract was then measured at 550 nm and recorded for further calculation of total
triterpenoid.
206
Chapter 7
Materials, methods and experimental procedures
7.3.8 High temperature-Gas chromatography mass spectrometry (HT-GCMS) of
heather extracts
Gas chromatography-mass spectrometry analysis was carried out using a Perkin
Elmer Clarus 500 chromatograph coupled with a Perkin Elmer Clarus 560S mass
spectrometer. A ZB5-HT (Phenomenex) capillary column (30 m x 0.25 mm I.D x
0.25 m film thickness, max temperature 430 oC) was used with constant carrier gas
flow rate of 1ml/min helium. Initial oven temperature was maintained at 60 oC for 1
minute, and then ramped at a rate of 8 oC/min to 360 oC. The oven temperature was
then held at 360 oC for 10 minutes before cooling. The injector was maintained at 350
o
C and the split ratio used was 10:1. Spectra were recorded in electron impact (EI)
mode at 70 eV scanning from 40-1200 amu in 1 second. The source temperature and
quadruple temperature were maintained at 350 oC. Peaks from the extracts were
initially identified by searching the NIST library (2008). Then some compounds were
further identified by comparing their retention time and spectrum with standard
material, the others were identified by comparing spectrum with published data of
previous researches. Fragmentation mechanisms were also studied to help confirm the
identification.
7.4 Extraction and fractionation of bracken
7.4.1 Supercritical carbon dioxide extraction of bracken fern
Supercritical CO2 extractions were carried out using a Thar SFE 500 system. A 3-step
extraction scheme was be used for bracken extraction. Liquid CO2, supercritical CO2
and then supercritical CO2 plus co-solvent were used sequentially to extract molecules
of increasing polarity.
207
Chapter 7
Materials, methods and experimental procedures
Firstly, the bracken sample was extracted using liquid CO2. Liquid CO2 extraction of
bracken (200g) was carried out at a pressure of 6.5 MPa and a temperature of 5 oC for
6 hours. CO2 flow rate was set to 40 g/min. The extract was washed out of the
evaporator using dichloromethane and then dried using a rotary evaporator to obtain a
yield. The residue from the first extract remained in the extractor and was extracted
using supercritical CO2 thereafter. The conditions of the first supercritical CO2
extraction were 35 MPa, 50 oC and a flow rate of 40g CO2/min for 4 hours. The
extract was again washed out of the evaporator using dichloromethane and then dried
using a rotary evaporator to obtain a yield.
Finally, the residue from the second extraction was extracted using supercritical CO2
and ethanol as a co-solvent. The conditions of the second supercritical CO2 extraction
were 35 MPa, 50 oC and a flow rate of 40 g CO2/min for 4 hours with addition of
ethanol at 10% of the CO2 flow, i.e. 4ml/min. During the extraction, the ethanol
extract solution was bled regularly from the evaporator to prevent overfilling. The
extract solution was finally combined and evaporated to dryness to afford a yield.
7.4.2 High temperature-Gas chromatography mass spectrometry (HT-GCMS)
method for bracken analysis
Samples prepared according to 7.4.1 were diluted at 20mg/ml dichloromethane for
GC-MS analysis. Gas chromatography-mass spectrometry analysis was carried out
using a Perkin Elmer Clarus 500 chromatograph coupled with a Perkin Elmer Clarus
560S mass spectrometer. A ZB5-HT (Phenomenex) capillary column (30 m x 0.25
mm I.D x 0.25 m film thickness, max temperature 430 oC) was used with constant
carrier gas flow rate of 1 ml/min helium. Initial oven temperature was maintained at
60 oC for 1 minute, and then ramped at a rate of 8 oC/min to 360 oC. The oven
temperature was then held at 360 oC for 10 minutes before cooling. The injector was
maintained at 350 oC and the split ratio used was 10:1. Spectra were recorded in
electron impact (EI) mode at 70 eV scanning from 40-1200 amu in 1 second. The
208
Chapter 7
Materials, methods and experimental procedures
source temperature and quadruple temperature were maintained at 350 oC. Peaks from
the extracts were initially identified by searching the NIST library (2008). Then some
compounds were further identified by comparing their retention time and spectrum
with standard material, the others were identified by comparing spectrum with
published data of previous researches. Fragmentation mechanisms were also studied
to help confirm the identification.
7.4.3 Methanol extraction of bracken for supercritical fluid chromatography
1 g of bracken powder was added to a sample bottle containing 5 ml of methanol.
This solution was then stirred for 4 hours at room temperature prior to centrifugation
at 3000 rpm for 5 minutes. The supernatant was decanted and transferred to a GC vial
for supercritical fluid chromatography.
7.4.4 Supercritical Fluid Chromatography of bracken extracts
An Investigator SFC system (Waters) coupled with integrated 2998 Photo Diode
Array detector and 2424 Evaporative Light Scattering detector was used to analyse
the bracken sample obtained in 7.4.3. A constant flow of 4ml/min supercritical CO2
along with methanol as organic modifier were used as the mobile phase. The system
was fitted with a Viridis (Waters) SFC silica column (25 cm×4.6 mm×5 m).
Analysis of the bracken extracts was carried using a gradient of increasing levels of
methanol in CO2. The percentage methanol was initially 10% rising from 10% to 25%
in 8 minutes and then holding at 25% for 1 minute. The methanol content was then
reduced back to 10% in 1 minute. The oven temperature was kept at 40 oC and back
pressure was set to 15.0 MPa. Compounds were identified by direct comparison of
retention time with standard compounds.
209
Chapter 7
Materials, methods and experimental procedures
210
Chapter 8
Conclusion and further work
Conclusion and further work
Chapter 8
211
Chapter 8
Conclusion and further work
8.1 Concluding remarks
8.1.1 General background
Heather (Calluna vulgaris) and bracken fern (Pteridium aquilinum) are two widely
distributed species which dominate the vegetation of large areas in many UK upland
regions particularly the North Yorkshire Moors. The North York Moors National Park
contains the largest patch of continuous heather moorland (50,100 ha) in England,
representing over 10% of the country’s resource. Heather is dominant on dry heath
vegetation which covers 26,500 ha (52.9%) and forms the main land cover on the
western, southern and central moors. Heather is also abundant on the wet heath (8,700
ha, 17.4%), which occurs on moister soils. Meanwhile, the total area of bracken fern
is estimated at 12,300 ha (24.6%) in North York Moors National Park and this figure
is increasing every year since bracken is seriously invading heather-dominated areas.
Currently, burning is the predominant method of managing moorland vegetation
especially heather. The most noticeable value of heather is as bird habitat for breeding
and feeding. To suit the bird’s different phases of growth, moorland needs to be burnt
to various heights to maximise the food availability since certain invertebrate species
are only associated with recently burnt or cut areas. On the other hand, bracken is a
major weed problem in Britain since its expansion is causing the loss of large areas of
semi-natural vegetation. Its vigorous growth makes it difficult to control and its dense
canopy shades out other species, reducing a site’s grazing or forestry potential.
Bracken can be controlled by herbicide use or mechanical means. At the moment, the
use of the herbicide Asulam (methyl N-[(4-aminophenyl)sulfonyl] carbamate)
accounts for most attempts at bracken control. It has a low toxicity for animals and
affects few other plants; however it still has some disadvantages such as high cost and
spray drift which can affect populations of other ferns. More seriously, Asulam spray
can cause sudden large-scale hydrological change that can lead to soil erosion before
212
Chapter 8
Conclusion and further work
colonisation by other species. For all these reasons, if any valuable compounds with
potential commercial application opportunities can be extracted from these two
currently wasted resources, harvesting can take place regularly at a specific time of
the year instead of burning and spraying herbicide. Within this study, heather and
bracken are demonstrated to be very good resources of some valuable chemicals
which can potentially be applied as herbal medicine, pharmaceutical drugs or drug
precursors.
8.1.2 Conclusions from heather study
Aerial parts of heather were harvested in October 2007, January 2008, March 2008
and August 2008 from Hole of Horcum, North York Moors National Park (Grid
reference SE 84558 93633) to be representative of heather in different seasons.
Heather samples were extracted using supercritical CO2 and supercritical CO2 with
ethanol entrainer. The highest yields in both dry plant and fresh plant were all
obtained in August flowering sample. Since supercritical carbon dioxide has been
described to have extracting ability similar to that of hexane, traditional hexane
Soxhlet extractions were also applied to give a yield comparison in heather. The
general yield variation trend of hexane extraction was the exactly the same as seen
with supercritical CO2 extraction. August heather expressed highest yield and winter
heather has the lowest. Since it was also reported that triterpenoid content is also
higher in summer flowering heather, August heather supercritical CO2 and
supercritical CO2 with ethanol entrainer extracts were fractionated by column
chromatography and fully analysed for their triterpenoid composition. Several
triterpenoids such as Friedelin and 13,27-cycloursan-3-one were discovered in these
supercritical CO2 extracts. The quantification of these new triterpenoids was also
accomplished by using oleanolic acid as a standard. Taraxerol and taraxerone which
have not been previously reported in heather leaves and stems were also found in
heather aerial part supercritical CO2 extracts. As with other triterpenoids, the
quantification of taraxerol and taraxerone were also achieved by using oleanolic acid
213
Chapter 8
Conclusion and further work
as standard. Taraxerol and taraxerone are both reported to have several biological
activities such anti-cancer and anti-inflammatory. Taraxerone showed in vitro
anti-leishmanial activity against promastigotes of Leishmania donovani and
anti-tumour activity on K562 leukemic cell line. Taraxerol was reported to have
significant activity against dermatophytes T. rubrum and T. mentagrophytes with MIC
values between 6.2 and 25 g/ml, additionally, taraxerol also express moderate
activity against Aspergillus niger.
Seasonal variation of total triterpenoid compounds in heather supercritical CO2 and
supercritical CO2 with ethanol entrainer extracts were also evaluated in this study. The
highest total triterpenoids content was found in August heather with 15400 g/g dry
plant and 12979 g/g fresh plant. March heather contains the second highest amount
of total triterpenoids at 11200g/g in dry plant and 9467g/g fresh plant. Winter
heather was found to have the lowest triterpenoid content. January heather only
obtains 4200g/g total triterpenoids in dry plant and 3620g/g in fresh plant.
October heather contains 8500g/g triterpenoids in dry plant and 7606g/g in fresh
plant. Heather total triterpenoids increase from winter to summer, then decrease again
in the autumn. Summer flowering heather was found to have the greatest amount of
total triterpenoid. The same trend was shown in heather supercritical CO2 with ethanol
entrainer extracts. However, these extracts were found to have slightly higher total
triterpenoid than supercritical CO2 extracts. Heather samples were also extracted
using hexane to give a comparison of total triterpenoid content with supercritical CO2
extracts. The seasonal variation of triterpenoid in hexane extracts was exactly the
same as seen with supercritical CO2 extracts. The greatest total triterpenoid content
was still found in August heather which is 17000 g/g dry plant and 14328 g/g fresh
plant.
The burning of heather takes place in spring every year since it is restricted by law
that burning can only take place between 1st October and 15th April. In March heather
sample, total triterpenoid content was evaluated as 11200 g/g dry plant in
214
Chapter 8
Conclusion and further work
supercritical CO2 extract and 12400 g/g dry plant in hexane extract. The total
triterpenoid content in March is approximately 88% of that of August heather sample.
Therefore, if harvesting can take place in March instead of burning, the frond will
already contain almost 90% of the highest triterpenoid content in the year. A large
amount of triterpenoid can be extracted if large-scale harvesting of heather can be
achieved at this time of the year.
Total phenolic content of heather was also analysed in this study. The lowest total
phenolic content was found in heather flowers, which is 2130 mg/100g dry plant.
Heather whole plant seem to have the lowest total phenolics in autumn and then keep
rising until reach a peak in spring sample at 5163 mg/100g dry plant. Afterwards the
total phenolic content drops slightly to 4201 mg/100g dry plant in summer when it is
flowering. Heather phenolic extracts were then evaluated for their anti-microbial
activity. Qualitative results showed that the seasonal variation of antimicrobial
activity of heather is in clear contrast to that of total phenolic content. This result
suggested that the antimicrobial activity does not necessarily correlate with total
phenolic content in heather. Quantitative tests were then carried out on phenolic
extracts at concentrations of 1000 ppm and 100 ppm. However, 1000 ppm heather
phenolic extract and 100 ppm phenolic extract only inhibited 9.83% and 9.06% of
total growth of lactobacillus respectively. These inhibition figures are very low
compared to other commonly used antimicrobial agents. Therefore, heather phenolic
extract at 1000 ppm and 100 ppm both have very weak antimicrobial activity and are
not sufficient to be applied as an antimicrobial agent even at these fairly high
concentrations.
8.1.3 Conclusions from bracken study
Samples in different bracken growth cycle stages were picked from the same site
(Kilburn White Horse, Sutton Bank, Grid reference SE 514 813) in different months
of a year. The first crozier sample was picked in 17th April 2009 and 19th May 2010,
215
Chapter 8
Conclusion and further work
followed by the 2nd step sample picked on 21st May 2009, 3rd sample picked in 27th
July 2009 and 4th sample picked on 12th October 2009. It was reported that the
carcinogenicity of young bracken shoots is much stronger than that of mature plant.
Due to this remarkable distinction of bracken crozier, liquid CO2 extraction, along
with supercritical CO2 extraction was employed in bracken crozier analysis to give a
more comprehensive understanding of its chemical content. Thus bracken crozier has
been extracted by liquid CO2, supercritical CO2 and supercritical CO2 with ethanol
entrainer sequentially. Two naturally occurring pterosins, pterosin B and pterosin F,
were successfully identified in bracken crozier liquid CO2 extract by comparing their
mass spectrum with published standard data. Other major components such as
-sitosterol and long chain primary alcohols were also detected in this extract at fairly
high concentration. All these identified components are still present in following
supercritical CO2 extract. More long chain alcohols with higher chain length were also
identified in supercritical CO2 extract, and their existence indicates that the extraction
ability of scCO2 is substantially higher than liquid CO2. With higher temperature and
pressure, supercritical CO2 is able to extract molecules with greater polarity and
higher molecular weight. This result indicates that the selectivity of CO2 depending
on operating conditions is achievable and successful. In bracken crozier supercritical
CO2 with ethanol entrainer extract, major peaks including pterosin B and-sitosterol
were still present, however pterosin F and long chain primary alcohols are no longer
present in this extract. Several new peaks appear in this extract due to the greater
polarity of scCO2 with ethanol entrainer. Some additional branched unsaturated
alcohols related to phytol were detected in this extract, and further sitosterol
derivatives also eluted immediately prior to the -sitosterol peak.
Bracken samples harvested in May 09, July 09 and October 09 were extracted by
supercritical CO2 and supercritical CO2 with ethanol entrainer sequentially. Most
interesting peaks such as pterosin B, pterosin F and -sitosterol all still present in May
bracken sample at very high concentration, however, the chemical composition in the
plant varied significantly after July. The most interesting change is in this fully
216
Chapter 8
Conclusion and further work
matured frond, pterosin B exist at very low concentration compared to younger frond,
and pterosin F was not even detectable. Both pterosin B and pterosin F no longer
appear in October senescing bracken extract. In both bracken supercritical CO2
extracts and supercritical CO2 with entrainer extracts, pterosin compounds exhibited a
clear reduction as the seasons progressed. Pterosin B and pterosin F disappeared in the
frond when it is growing towards its maturity; however-sitosterol is present in
bracken frond at every growth stage.
Seasonal variation of pterosins compounds and other components are also evaluated
in this study. The highest total water content of bracken was found in crozier at
73.40%, and then it decreased progressively during the growth cycle until it reached
the lowest point in October senescing bracken at 44.80%. The total yields of
supercritical carbon dioxide extract in both dry plant and fresh plant increases from
crozier to over-matured plant. Due to lack of pterosin standards, the quantification of
pterosin compounds was not accomplished. However, octadecane was introduced as
an internal standard to give a direct comparison of the relative level of pterosins and
octadecane in peak height/area. Result shows that pterosin B and pterosin F decreased
dramatically when plant grows towards its maturity. Geographical difference was also
found to cause significant variation in the level of pterosin compounds. Bracken
crozier samples picked from Penrallt in North Wales in May 2010 was extracted using
supercritical CO2 under identical conditions. It was found that Welsh bracken contains
much less pterosin B and pterosin F compared to Yorkshire sample. This distinct
difference may be due to different pH value of the soil in Kilburn and Wales.
Recently, pterosins are discovered to have anti-diabetic and anti-obesity activities.
Totally 84 different pterosins or pterosin related compounds were synthesized or
extracted from plants and have been tested in vitro and in vivo for their anti-diabetic
and anti-obesity activities. 15 of them expressed very strong biological activities
against type 1 and type 2 diabetes and obesity, and among these 15 pterosins, four of
them are naturally occurring in bracken fern and some other natural plant materials.
217
Chapter 8
Conclusion and further work
These pterosin compounds were indicated to significantly lower blood glucose levels
in STZ-induced diabetic mice, and also dramatically enhanced the insulin sensitivity
and glucose consumption in vitro. The pterosin compounds may activate AMPK
(AMP-activated protein kinase), which in turn regulates insulin regulation of
carbohydrate and fatty acid metabolism. These compounds can therefore be
considered as potential anti-diabetic and anti-obesity agents. Further results also
indicated that these pteorsin compounds can significantly reduce serum lipids and
exhibited anti-obesity effects in high fat-diet fed mice.
Nowadays, most pterosin compounds are synthesized through a series of complex
processes. For example, a typical method to synthesize pterosin F is mixing pterosin
B and triphenylphosphine in carbon tetrachloride and reflux for 3 hours. The starting
material pterosin B was obtained from pterosin O by demethylation with 48%
hydrobromic acid in acetic acid. Moreover, pterosin O was afforded through catalytic
hydrogenation with 5% Pd on carbon in ethyl acetate from an intermediate indandione,
and this key intermediate was formed through a Friedel-Crafts bisacylation of the
methyl ether of 2-(2,6-dimethylpheny1) ethanol with methylmalonyl chloride. The
synthesis processes are very complicated and will generate large amounts of chemical
waste and solid waste. However, in my study, pterosin F and pterosin B were
demonstrated to be naturally occurring in bracken liquid CO2 and supercritical CO2
extracts at fairly high concentration. Therefore, it is clear that naturally occurring
valuable pterosins can be extracted from bracken by using environmental benign
technologies rather than synthesized through complex processes. This study will not
only add value to this troublesome natural plant, but will also make some contribution
to the green chemistry application in this research field.
To the best of my knowledge, bracken frond has never been extracted by liquid or
supercritical carbon dioxide before. Previous common extraction methods of bracken
were limited to water extraction or methanolic extraction. In my study, liquid CO2 and
supercritical CO2 were indicated to be capable to extract at least two natural pterosins
218
Chapter 8
Conclusion and further work
from bracken crozier and pre-mature frond with high concentration. Although pterosin
B and F were not identified to have certain anti-diabetic or anti-obesity activities in
previous research, these pterosin compounds, especially pterosin F, have the
possibility to be used as precursors to other more effective pterosins because of its
active chlorinated side chain.
To conclude, heather and bracken can both be good potential resources of herbal
medicine or pharmaceutical drugs and precursors. Heather was proved to have high
content of triterpenoid compounds which have been reported for their biological
activities such as anticancer and liver protecting property. Bracken was proved to
contain high content of natural occurring pterosin B and pterosin F which can be
converted to other pterosins which have been proven for their anti-diabetic and
anti-obesity activities. All these high value target chemicals can be extracted by
environmental benign liquid carbon dioxide or supercritical dioxide.
8.2 Future work and publications
Heather flower supercritical CO2 extract and supercritical CO2 with ethanol entrainer
extracts were fractionated by column chromatography and 14 or 11 final fractions
were obtained respectively, however, only a few of them which contain triterpenoid
compounds were selected to be fully analysed. Further analysis can be carried out on
the more polar fractions to look for more interesting components with valuable
biological activities.
It was clear that bracken crozier and pre-mature frond contains the highest amount of
pterosin compounds. If young bracken frond can be harvested regularly, it will be able
to afford larger scale of liquid and supercritical CO2 extractions. Therefore higher
amount of pterosin compounds especially pterosin F which is more likely to be
applied as a pharmaceutical precursor can be extracted. Thereafter, an isolation and
219
Chapter 8
Conclusion and further work
purification of pterosin compounds can be achieved by supercritical fluid
chromatography which is a proven purification application used in the pharmaceutical
industry. Pure pterosin compounds can be then analysed individually for biological
activities, or converted to other pterosins which have been proven to be anti-diabetic
and anti-obesity.
This study has revealed new aspects of the chemistry of heather and bracken as well
as demonstrating the commercial potential for harvesting this material. It is intended
that the identification of new triterpenoids in heather and the extraction of bracken
using liquid and supercritical CO2 will be published as two separate papers in peer
reviewed journals.
220
List of references
List of references
1
R. J. Pakeman, P. D. Hulme, L. Torvell, J. M. Fisher, Biological Conservation
2003 114 389-400
2
A. J. Hunt, The Extraction of High Value Chemicals from Heather (Calluna
Vulgaris). PhD Thesis, University of York, York, 2006
3
R. J. Pakeman, R. H. Marrs, Biological Conservation 1992 62 101-114
4
M. Shahin, B. L. Smith, A. S. Prakash, Mutation Research 1999 443 69–79
5
R. J. Pakeman, Applied Geography 1996 16 1 65-86
6
K. Yamada, M. Ojika, H. Kigoshi, Natural Product Reports 2007 24 798–813
7
K. Yoshihira, M. Fukuoda, M. Kuroyanagi, S. Natori, M. Umeda, T. Morohoshi,
M. Enomoto, M. Saito, Chem. Pharm. Bull. 1978 26 2346
8
P. Sengufta, M. Sen, S. K. Niyogi, S.C. Pakrash, E. Ali, Phytochemistry 1976
15 995
9
H. Sheridan, N. Frankish, R. Farrell, Planta Medica 1999 65 3 271-272
10
J. B. Manley, P. T. Anastas, B. W. Cue Jr., Journal of Cleaner Production 2008
16 743-750
11
P. Anastas, J. Warner, Green Chemistry: Theory and Practice Oxford
University Press 1998
12
A. K. Mohanty, M. Misra,L. T. Drzal, Journal of Polymers and the
Environment 2002 10 19-26
13
L. Wang , C. L. Weller, V. L. Schlegel, T. P. Carr, S. L. Cuppett, Bioresource
Technology 2008 99 1373–1382
14
J. Liu Journal of Ethnopharmacology 2005 100 92–94
15
Y. Ikeda, A. Murakami, H. Ohigashi, Mol. Nutr. Food Res. 2008 52 26-42
16
L. Novotny, A. Vachalkova, D. Biggs, Minireview. Neoplasma 2001 48
241-246
17
S. Rao, G. Muralikrishna, J. Agric. Food Chem. 2002 50 889-892
18
I. Lambropoulos, I. G. Roussis, Eur. J. Lipid Sci. Technol. 2007 109 623–628
221
List of references
19
M. Friedman, J. Agric. Food Chem.1997 45 1523-1540
20
B. Wang, Z. H. Jiang, Chin. Pharm. J.1992 27 393
21
A. R. Waladkhani, M. R. Clemens, Int. J. Mol. Med.1998 1 747-753
22
S. Balanehru, B. Nagarajan, Biochem. Int. 1991 24 981-990
23
S. Balanehru, B. Nagarajan, Biochem. Int. 1992 28 735-744
24
Y. G. Liu, Chemical Abs. 1986 106 90209
25
N. 1986, T. Honda, H. J. Finlay, Cancer Res. 1998 58 717–723
26
M. B. Gupta, T. N. Bhalla, G. P. Gupta, C. R. Mitra, K. P. Bhargava, 1969 6
67-70
27
G. B. Singh, S. Singh, S. Bani, B. D. Gupta, S. K. Banerjee, J. Pharm.
Pharmacol. 1992 44 456–458
28
H. Ohigashi, H. Takamura, K. Koshimizu, H. Tokuda, Y. Ito, Cancer Lett. 1986
30 143–151
29
H. Tokuda, H. Ohigash, K. Koshimizu, Y. Ito, Cancer Lett. 1986 33 279–285
30
M. Niikawa, H. Hayashi, T. Sato, H. Nagase, H. Kito, Mutat. Res. 1993 319 1-9
31
J. Liu, J. Ethnopharmacol. 1995 40 57-58
32
J. Liu, Y. Liu, C. D. Klaassen, J. Ethnopharmacol. 1994 42 183-191
33
J. Liu, Y. Liu, Q. Mao, C. D. Klaassen, Fundam. Appl. Toxicol. 1994 22 34-40
34
V. R. Yadav, S. Prasad, B. Sung, R. Kannappan, B. B. Aggarwal, Toxins 2010 2
2428-2466
35
A. Fournet,, A. Angelo, V. Munoz, F. Roblot, J. Ethnopharmacol. 1992 37
159-164
36
T. H. Beveridge, T. S. Li, J. C. J. Agric. Food Chem. 2002 50 744-750
37
R. Kakuda, T. Iijima, Y. Yaoita, K. Machida, M. Kikuchi, Phytochemistry 2002
59 791-794
38
M. B. C. Gallo, M. J. Sarachine, International Journal of Biomedical and
Pharmaceutical Sciences, 2009 3 46-66
39
M. A. F. Jalal, D. J. Read, E. Haslam, Phytochemistry, 1982 21 6 1397-1401
40
M. P. Kahkonen, A. I. Hopia. H. J. Vuorela, J. Rauha, K. Pihlaja, T. S. Kujala,
M. Heinonen, J. Agric. Food Chem., 1999 47 3954-3962
222
List of references
41
H. Liu, N. Qiu, H. Ding, R. Yao, Food Research International 2008 41 363-370
42
Council for Agriculture Science and Technology, Nutraceuticals for Health
Promotion And Disease Prevention 2003
43
M. Corrales, J. H. Han, B. Tauscher, International Journal of Food Science and
Technology 2009 44 425-433
44
I. Hirono, Y. Kono, K. Takahashi et. al, Vet. Rec. 1984 115 375–378
45
D. M. X. Donelly, J. O'reilly, J. Polonsky, M. H. Sheridan, J. Chem. Soc. Chem.
Commun.1983 615
46
U. F. Castillo, A. L. Wilkins, D. R. Lauren, B. L. Smith, N. R. Towers, M. E.
Alonso-Amelot, R. Jaimes-Espinosa, Phytochemistry, 1997 44 901-906
47
A. Kobayashi, K. Koshimizu Agric. Biol. Chem. 1980 44 393
48
A. Kobayashi, H. Egawa, K. Koshimizu T. Mitsui, Agric. Biol. Chem. 1975 39
1851
49
P. Sengufta, M. Sen, S. K. Niyogi, S. C. Pakrashi, E. Ali, Phytochemistry, 1976
15 995
50
F. L. Hsu, S. H. Liu, B. J. Uang, Use of Pterosin Compounds for Treating
Diabetes and Obesity World Intellectual Property Organization July 2010
WO2010085811A2
51
K. E. Ng, T. C. McMorris, Can. J. Chem. 1984 62 1945-1953
52
W. K. Modey, D. A. Mulholland, M. W. Raynor, Phytochemical Analysis 1996
7 1-15
53
K. Muffler, Downstream Processing in Marine Biotechnology 2005 p.79
54
http://sunny.vemt.bme.hu/sfe/angol/main.html, Group of Supercritical Fluid
Extraction, Budapest Uuniversity of Technology And Economics
55
Thermophysical Properties of Fluid Systems, NIST Standard Reference Data
Nist Webbook
56
R. Murga, R. Ruiz, S. Beltran, J. L. Cabezas, J. Agric. Food Chem. 2000 48
3408-3412
57
L. Wang, C. L. Weller, Trends in Food Science & Technology 2006 17
300–312
223
List of references
58
M. T. Tena, M. Valcarcel, Anal. Chem. 1997 69 521-526
59
L. Castro, L. E Garcia-Ayuso, Analytica Chimica Acta, 1998 369 1–10
60
R. Zarnowski, Y. Suzuki, Journal of Food Composition and Analysis 2004 17
649–664
61
P. K. Mamidipally, S. X. Liu, European Journal of Lipid Science and
Technology 2004 106 122–125
62
D. J. Adams, P. J Dyson, S. J. Taverner, Chemistry in alternative reaction
media, Wiley, Chichester, 2004
63
H. Li, L. Pordesimo, J. Weiss, Food Research Internationa, 2004 37 731–738
64
Y. Zhao et al., LC GC Europe 2004 174 224-2
65
R. D. Pancost, M. Baas, B. V. Geel, J. S. S. Damste, Organic Geochemistry
2002 33 675-690
66
J. Cook, H. Chable-Rabinovitch, M. Rigaud, C. Delace, A. Simon, A. J. Chulia,
A. Najid, FEBS Lett., 1992 299 213-217
67
A. A. Khan and P. E. Kolattukudy, Biochemical and Biophysical Research
Communications, 1974 61 1379-1386
68
R. W. Lewis, Phytochemistry, 1972 11 417-419
69
H. Schaller, Progress in Lipid Research, 2003 42 163-175
70
J. P. Kutney, G. Eigendorf and I. H. Rogers, Tetrahedron Letters, 1969 25
3753-3766
71
B. Moulisha, M. Bikash, P. Partha, G. A. Kumar, B. Sukdeb, H. P. Kanti,
Tropical Journal of Pharmaceutical Research, 2009 8 127-131
72
B. Aguilar-Guadarrama, V. Navarro, I. Leon-Rivera, M. Y. Rios, Natural
Product Research, 2009 23 1559-1565
73
Y. Ding, C. Liang, J. H. Kim, Y. Lee, J. Hyun, H. Kang, J. Kim, B. S. Min, Y.
H. Kim, Bioorganic & Medicinal Chemistry Letters 2010 20 1528-1531
74
L. He, X. Yue, Y. Chen, Chinese Journal of Applied Chemistry, 1996 13 67-69
75
V. J. Ndlebe, N. R. Crouch, D. A. Mulholland, Phytochemistry Letters, 2008 1
11-17
224
List of references
76
D. H. R. Barton and P. de Mayo Journal of the Chemical Society, 1953
2178-2181
77
M. Jose, U. Ferreira, J. R. Ascenso and O. S. Tavares, Journal of Natural
Products, 1995 58 275-279
78
G. F. Aragao et al., Pharmaceutical Biology, 2007 45 343-349
79
S. Johann, C. Soldi, J. P. Lyon, M. G. Pizzolatti, M. A. Resende, Lett. Appl.
Microbiol 2007 45 148-153
80
A. H. Murray, G. R. Iason, C. Stewart, Journal of Chemical Ecology 1996 22
1493-1504
81
L. Bravo, Nutrition Reviews 1999 56 317–333
82
A. Molero Gomez, C. Pereyra Lopez, E. Martinez de la Ossa, The Chemical
Engineering Journal 1996 61 227-231
83
A. J. Hunt, E. H. K. Sin, R. Marriott, J. H. Clark Chemsuschem 2010 3 306-322
84
F. Bonadies, G. Borzacchiello, S. Dezzi, R. Nicoletti, S. Roperto Rapid
Communications In Mass Spectrometry 2004 18 825-828
85
M. Saito, M. Umeda, M. Enomoto, Y. Hatanaka, S. Natori, Z. Yoshihira, M.
Fukuoda, M. Kuroyanagi, Experientia 1975 31 827-831
86
C. G. Jones, R. D. Firn, Biochemical Systematica and Ecology, 1979 7 95-101
87
K. Saito, T. Nagao, M. Matoba, K. Koyama, S. Natori, T. Murakami, Y. Saiki,
Phytochemistry 1989 28 1605-1611
88
K. Koyama, S. Takatsuki, S. Natori, Phytochemistry 1991 30 2080-2082
89
H. G. Preuss, C. Marcusen, J. Regan, I.W. Klimberg, T. A. Welebir, W.A. Jones,
International Urology and Nephrology 2001 33 217–225
90
M. E. Alonso-Amelot, S. Rodulfo-Baechler, Biochemical Systematics and
Ecology 1995 23 709-716
91
L. S. Rasmussen, Ptaquiloside-An Environmental Harzard? Ph.D Thesis, The
Royal Veterinary and Agricultural University, Demark 2003
92
G. J. D. Kirk, P. H. Bellamy, R. M. Lark, Global change biology 2010 16 3111
-3119
225
List of references
93
H. Yamashiro, H. Oku, H. Higa, I. Chinen, K. Sakai, Comp. Biochem. Physiol.
B., 1999 122 397-407
94
A. L. Waterhouse, Current Protocols in Food Analytical Chemistry 2002
I1.1.1-I1.1.8
95
W. Wang, D. Shang, L. Wen, Edible Fungi of China 2006 25 30-32
226